This PDF file contains the front matter associated with SPIE Proceedings Volume 12294, including the Title Page, Copyright information, Table of Contents, and Conference Committee Page.

UHV line project has the characteristics of "long lines with many points", and the foundation type is greatly affected by the complex geology along the line. In order to make the line foundation type suitable for the geology along the line and better economy, Taking Nanchang Changsha UHV transmission line project as the background, the economic analysis and comparison of four foundation types along the line are carried out, in order to provide reference value for similar projects.

High-power electromagnetic pulse (HPEM), which has steep rising front, high radiation power and wide spectrum, can coupling into the electronic equipment cabinet to damage the internal electronic equipment, such as computers. Coupling characteristics of HPEM on electronic equipment cabinet worth studying. Based on FDTD method, the coupling model of HPEM into electronic cabinet is established, and the sub-grid generation technology is used to deal with the small aperture on the cabinet, so as to realize the accurate simulation of the coupling process of HPEM. Through the simulation, conclusions of cabinet resonance effect and aperture gap enhancement effect are found.

In order to monitor the working condition of vacuum pump in real time and find the fault equipment, implemented a vacuum pump fault monitoring system. The system includes data acquisition, fault diagnosis, information display and other functions. It uses multi-thread and producer-consumer model to realize thread-safe data flow, and has a graphical interface. The system collects data, calculates its wavelet packet energy, selects the feature by mean decrease impurity method, and then uses the improved random forest algorithm for diagnosis after optimizing the weight of decision tree by out of bag estimation. Through testing, the accuracy of the improved model is more than 95%, and the system can realize real-time judgment, which has achieved good effect in practical application.

The land power system has a great number of long-distance lines, and the connectivity between power grids is inadequate, resulting in negative damping of the power system. The PSS+AVR excitation control approach has been widely utilized in the single machine infinite bus power grid. The distribution line was short in the early ship power system, and the power system stabilizer was rarely employed. High-power propulsion motors have become one of the most important burdens on ships as a result of the widespread usage of marine integrated electric propulsion systems. we use pss2B to the excitation management of a zonal distribution power system, taking into account the ship power station's significant nonlinear characteristics. The comparison results suggest that pss2B may be employed in ship power systems and has a positive effect on load disturbance under various working situations.

The equipment and device protocols in the power monitoring system are difficult to unify, and the existing power monitoring system structure system analysis protocol takes a long time. In order to improve the system operation efficiency, a power monitoring system architecture based on multi-protocol fusion is designed. Establish a power monitoring system model through the power and state change equations; coordinate the data analysis in the system and the caching and reading functions of key logic to ensure multi-protocol fusion and conversion of power monitoring data; reset the power monitoring architecture based on the multi-protocol fusion method, establish a multi-objective optimization model, set its constraints, and obtain a new power monitoring system architecture. Taking the user scale as a variable, comparing the response time of four different power monitoring systems, it can be seen from the experimental data that the time required for this architecture is the minimum value under the same conditions under five different operations. The optimization effect is the best, which can realize the fast response of the power monitoring system.

Electric Motors play a vital role in various aspects of production and life. Once a motor fails, the loss will be incalculable, and the state of motor bearings will directly affect the reliability of motor operation. Due to the high failure rate of motor bearings and the difficulty of accurately identifying fault types, a fault diagnosis method based on Wavelet Singular Entropy (SEM) and Support Vector Machine (SVM) was proposed. The fault was implanted into the motor bearing by electro-discharge machining (EDM), and the vibration signal samples of 4 different state types were measured. The characteristics of the four state type vibration signals are extracted with wallet singular entropy, and the state classification of the data after feature extraction is carried out by the support vector machine classifier. The experimental research shows that the fault diagnosis method of motor bearing based on wavelet singular entropy and support vector machine can greatly improve the fault diagnosis accuracy.

The existing security alarm system is difficult to accurately judge the effective alarm data, resulting in a large amount of redundant data in the alarm system database. In order to alleviate the operation pressure of the alarm system, a power monitoring network security threat alarm system based on attribute attack graph is designed. In the system hardware, the CPU power supply and network interface are designed. In the system software, the visual traceability analysis of security threat data is carried out. Based on the attribute attack graph, the topology algorithm is used as the analysis tool of attack nodes, and the weight index of information similarity is established. The binary attack node system in the system is established by atomic attack node and attribute attack node to judge the similarity level of different alarm information. In the experiment, the alarm aggregation rate and system reliability of four different alarm systems are tested. The experimental data show that the system can remove more redundant data without losing alarm data, which shows that the system has better performance.

Voltage sag can result in petrochemical enterprise motor running back in the process of production and process abort, lower the quality of power supply, cause huge economic losses. In view of the problem that the current voltage tolerance evaluation of petrochemical plants mainly adopts the traditional sensitive equipment tolerance curve, which cannot accurately describe the voltage sag tolerance of sensitive loads, this paper proposes a voltage sag simulation and evaluation method based on nameplate data through theoretical derivation and simulation verification. First of all to establish mathematical model of induction motor based on nameplate data to calculate the model parameters of the motor; Then, using the mechanical characteristic formula of induction motor and the mechanical characteristic formula of fan or pump load, the time of transient arrival boundary removal corresponding to voltage sag depth under different loads was deduced. Finally, a simulation model was built in Digsilent software to verify the correctness of theoretical analysis.

In the security agent communication of power monitoring system, because the front-end machine corresponds to multiple communication equipment, when one front-end machine fails, the load of the front-end machine system is unbalanced, resulting in system paralysis. Therefore, based on Docker technology, a safe load method of power monitoring system is designed. The security agent communication model of the power monitoring system is established. Based on the model, the Docker container is allocated to the front mobile state of the master station, combined with the telescopic group service to balance the load of the front-end computer system. According to the network configuration information of the Docker container, the communication connection between the acquisition equipment and the master station is established, the computer processes the access request, and sets the routing strategy according to multiple constraints, The information transmission channel and encrypted transmission channel are established by using the routing strategy to complete the design of safe load method of power monitoring system. The experimental results show that using the proposed safe load method of power monitoring system based on Docker technology, the load of front-end computer system is more balanced, and the load rate is between 40 ~ 50%, which can quickly process the communication and improve the stability and security of the system.

The sudden three-phase short circuit of synchronous generator is the most serious fault in power system, which may have a serious impact on the motor itself and related electrical equipment. It is of great significance to study. Under the D-P coordinate system, the mathematical model and dynamic equivalent circuit of synchronous generator are constructed. By using the powerful functions of Matlab/Simulink, a simulation model of a sudden three-phase short circuit at the generator end of the synchronous generator is constructed, and the physical quantities of the synchronous generator are simulated and studied during the short circuit. Through the comparative analysis of theory and simulation, the physical quantities of the synchronous generator have great impact and oscillation during the transient process of sudden short circuit, and finally stabilized in the state before the short circuit. The simulation results are consistent with the theoretical analysis. This method can also be used to study some dynamic processes of synchronous generators, which can provide the necessary theoretical basis for the optimal design of the motor.

In order to strike the target with high precision, the beam pointing accuracy of laser weapon needs to reach μrad level. The hysteresis, creep and high frequency non-linearity of piezoelectric ceramics (PZT) are the main factors that affect the system's pointing accuracy. In order to improve the dynamic control accuracy of fast steering mirror (FSM), the motion principle of FSM is introduced. The electromechanical model and the mathematical model of control object are proposed. The Smith predictive control algorithm is used to design control parameters and to validate the control algorithm by simulation.

Aiming at the problems of low efficiency, large delay and low throughput in LDPC decoding of OAI projects in current software defined radio platforms, this paper proposes a multi segment parallel decoding acceleration scheme for LDPC based on thread pool, and compares the LDPC decoding time and throughput between the architecture of this paper and the original OAI architecture under different LDPC segment numbers. Compared with the original OAI architecture, the architecture in this paper reduces the time required for LDPC decoding by 90.33% and increases the throughput by 145.71%, it improves the performance and practicability of the OAI project.

Based on the current characteristics of street lighting, the sensor data is extracted from the existing smart lighting devices, and the lighting failure prediction model of pigeon optimal BP network is built. The algorithm introduces magnetic navigation and landmark navigation in pigeon-inspired optimization into BP network, which solves the drawbacks of slow convergence and easy falling into local optimum of traditional network. The simulation results show that the algorithm can be applied to the prediction and diagnosis of urban street light illumination faults.

With the continuous progress of science and technology, more and more solar cells are used in the field of green power generation. Silicon solar cells, which are widely used as solar cells, dominate the solar cell market. Along with this comes a lot of waste silicon solar cells reaching the end of their useful life. In this essay, experimental data were collected by changing the time and solubility of solution at the same room temperature through the control variable method. The most efficient recovery conditions are obtained based on the existing recycle methods. This efficient recycling method can be applied to a large number of existing abandoned silicon solar cells to improve the recycling efficiency of enterprises and increase economic benefits. At the same time, it can also maintain efficient recycling and utilization of resources, reduce environmental damage and promote technological progress.

In order to study the sensor configuration of fire monitoring system on fan engine room. First of all, a mathematical model for fire calculation is established. Secondly, the data of heat release rate and the change rule of the temperature, heat flow and smoke height in fan engine room after fire are analyzed, according to the characteristics of heat flow field. Then, the detector is configured and the model of fan engine room is built. Finally, a real ignition test was carried out in the key hidden spot of fire, and different detection techniques were compared. The rationality of sensor configuration is verified and the response time of different detectors is obtained through experiments, which provides data support for realizing the best fan fire safety protection system.

Power monitoring system can monitor the operation of power system, but there are still defects in analysis and early warning. Most of the current safety analysis methods rely on a single dimension for anomaly analysis and early warning, resulting in large deviations in analysis results and low alarm reduction rate. In order to solve the above problems, the security analysis method of power monitoring system based on big data technology is studied. Spark technology is used to establish a big data framework for security analysis of power monitoring system, and the communication link data of power monitoring system is analyzed from the perspective of data statistics and abnormal characteristics. Deep neural network in machine learning is used to analyze the abnormal analysis results and realize effective power monitoring and early warning. Through the comparison test of an example, it is verified that the prediction accuracy of this security analysis method is higher than 94.8%, and the alarm reduction rate is significantly improved, which has good practical application value.

Aiming at the problem of low accuracy of single-phase grounding fault line selection in distribution network, a fault line selection method of distribution network based on twice regulating zero-sequence voltage is proposed in this paper. This method use the characteristics that the change law of zero sequence current normal line fault remains unchanged before and after grounding fault occurs, and it always changes in proportion to the system zero sequence voltage, while the change law of zero sequence current of fault feeder changes. By constructing the characteristic function, then regulate zerosequence voltage twice before and after fault occur, and calculated value of the characteristic function of each line, the line with the largest characteristic function value can be determined as a faulty line. This method eliminates the influence of three-phase admittance asymmetry, and has strong line selection ability of high resistance fault. The results of PSCAD simulation show that this method can correctly select the line under different fault conditions, and still has high accuracy in case of high resistance grounding fault.

Long-term partial discharge (PD) can lead to equipment insulation failure, resulting in huge economic losses, and PD signals are difficult to capture. In order to improve the reliability of partial discharge, this paper proposes a PD classification algorithm based on bidirectional long short-term memory network and attention mechanism (BI-LSTM-Attention). This method takes the PD signal as input, uses the BI-LSTM network to extract the features of the sequence before and after. Combined with the attention mechanism to adjust the feature weight, and finally through the sigmoid activation function to classify the results into two categories, get the PD recognition results. Tested on the ENET public dataset, the experimental results show that compared with the traditional long short-term memory network (LSTM), the loss rate and accuracy rate of the classification results are better, indicating that the proposed model based on BI-LSTM-Attention is more effective for classification tasks, which reflects the superiority of the algorithm.

In testing gun muzzle blast fields, deploying distributed wireless sensor networks is a flexible and efficient data acquisition method. However, the data transmission characteristics with the high sampling rate of single nodes and the high concurrency of multiple nodes lead to an excessively large overall data volume, which exceeds the existing wireless bandwidth carrying capacity. Therefore, this paper introduces distributed compressed sensing into the shock wave field test to reduce the overall transmission data volume by downsampling the sensor sources to meet the bandwidth requirements. Firstly, the paper analyses the signal-to-signal correlation of multiple shockwave signals within the network to meet the DCS down sampling constraints. Secondly, the performance of typical reconstruction algorithms in single and joint reconstruction is compared. Finally, the feasibility of applying DCS to shockwave field testing is verified, which provides a new idea for the data acquisition method of distributed wireless sensing networks.

This design mainly uses the equivalent area algorithm to calculate the inverter to generate the SPWM waveform, and keep the output waveform undistorted. At the same time, the ADC0809 is controlled by the single-chip microcomputer to collect and output the voltage value and display it on the digital tube. The main requirements of the design are low cost, small size, no need to adjust external components, and simple interface. The generation of SPWM is generated by the singlechip ATMEGA8 according to the algorithm. After isolation, drive and amplify, and finally filter the output, the required sine waveform was obtained. This subject is very practical and is used in many fields, such as: used in AC motor speed control systems, ships, aerospace, post and telecommunications, military equipment, transportation facilities, instrumentation, industrial equipment.

A novel method is proposed in this paper for extracting dielectric constants of non-magnetic materials and deriving material thickness. The extraction process is independent of thickness and does not have ambiguous value. Then the thickness derivation is conveniently based on the calculated dielectric constants. The proposed method is validated by simulation data and achieves reliable dielectric constants and material thickness value. It is verified in the waveguide measurement system and also suitable for coaxial line and free space measurement. This method is appropriate for materials whose properties are unknown or thickness cannot be exactly measured.

Discrete sampling is broadly utilized in the measurement system of wide-band partial discharge (PD) , and collection parameters of discrete sampling have a significant impact on the PD pulse sequence. Stationary noise originating from vertical resolution of oscilloscope is a feature of sampling signal, which always exists when digital acquisition system is used. In the wide band partial discharge measurement system, sampling frequency F_s, sampling time T_s, sampling bits and other parameters will lead to steady noise changes from oscilloscope, which can be quantified by signal to noise ratio (SNR). With the development of wide-band PD measurement system, clustering graph is the tool that can be used to analyze PD pulse in actual time, and apply clustering technology to PD source separation. By drawing classification diagram with equivalent time T_eq and equivalent frequency F_eq, characteristics of partial discharge pulses can be extracted and clustered, so as to realize the classification of sources. In the paper, a series of simulation analysis and measurements in laboratory are used to evaluate the impact of sampling parameters on T_eq and F_eq results. The results show that clustering outcomes are greatly influenced by the number of sampling bits, and the impact on the classification map will be reduced by the appropriate number of sampling bits.

The core function goal of virtual power plant (VPP) is to integrate all kinds of flexible resources driven by value, participate in all kinds of transactions in power wholesale market, provide capacity and auxiliary services for power grid operation, improve the economy and reliability of power system, and promote the efficient optimization and integration of renewable energy. However, there is no effective research on the income distribution mechanism of virtual power plant. This paper constructs a set of index system to measure the valuable contribution of each aggregated resource, and distributes the income reasonably according to the final valuable contribution of each aggregated resource. The actual example shows that this income distribution mechanism can effectively balance the interests of all parties and achieve a win-win situation of the system.

The reform of transmission and distribution electricity price has a great impact on the investment management and benefit recovery mode of power grid enterprises. As the object of supervision, power grid enterprises pay more attention to investment benefits. Firstly, in the dynamic benefit evaluation of power grid investment, the index scores are determined according to the index measurement standards, and the weight of the evaluation index system is determined according to the combination of subject and object. Secondly, in the calculation of accurate distribution of power grid investment, the 0-1 planning method is mainly used to calculate the specific project, establish the objective function and constraints, and then use lingo software to apply the established model.

Currently, all of first-round pilot provinces in China have conducted trial operations to verify the effectiveness and correctness of provincial power spot market rules. Although, different models and mechanisms are adopted according to the realistic situation in different provinces, evaluation index with general applicability is required to indicate and evaluate the progress of different provinces. Based on worldwide evaluation experience of power spot market, the provincial spot market evaluation index framework is proposed, which takes provincial realistic situation into consideration. In addition, the key indicators of provincial spot market evaluation are designed based on different application scenarios, providing reference for further improving the rules and mechanism of power spot market in China.

A day-ahead optimal economic dispatch method for industrial users based on shared energy storage power stations is proposed. Firstly, the concept of sharing energy-storage station is proposed and its business operation mode is analyzed. Then the shared energy storage power station is applied to the economic optimal scheduling of industrial users. By coordinating the power of charging and discharging of each user using the shared energy storage power station, the daily operation cost of the user group is optimized. Finally, it is concluded that the introduction of the sharing energy-storage station can significantly reduce the daily operation cost of the user group, and further study is conducted on the relationship between the income of service fee, the static investment recovery period and the investment return rate and the service fee pricing of the sharing energy-storage station.

China's power market and carbon market development have been accelerated to achieve the goal of decarbonisation with cost optimization. It is important to design a power market that takes into consideration the interactions with carbon trading. In this paper, the mutual influence of the power market and carbon market is analyzed. The two markets co-ordinately function to reduce carbon emission. Then, the international practices of power market mechanism that facilitate carbon market effectiveness are studied to bring forward suggestions for China's power market design. By strengthening the synergy mechanism between the carbon market and the power market, the role of the market mechanism in optimizing energy allocation and climate governance can be maximized.

At present, the profound changes in the international and domestic situation may cause continuous weak consumption and investment, will bring more uncertain impact on the power grid enterprise management, at the same time in the competition, supervision and performance pressure intensified, it is necessary to improve the utilization efficiency of the distribution network equipment assets, so it must be around the distribution network equipment assets management decision optimization support. This paper combines the concept of grid, referring to the existing results, considers applicability, improvement, accessibility and other factors, constructs multi-attribute evaluation index system, calculates hierarchical analysis, and divides different grids with power distribution system planning theory and management decision under the new situation.

In recent years, with the continuous expansion of transmission line project investment scale, the increasing national investment control of power grid enterprises and the change of business mode of power grid enterprises, higher requirements have been put forward for the lean control level of transmission line project cost. Therefore, combined with the requirements of transmission line engineering cost management and control, this paper constructs an index system for monitoring project cost variation based on big data analysis, and applies the normal distribution theory to construct corresponding monitoring technology methods. The method can provide technical guidance for improving the cost control level of transmission line engineering.

The development of the physical information fusion technology, the explosive growth of data, it marks the power system has entered the era of big data, thus increasing the information platform of power system, dealing with the difficulty of data storage, management and analysis. And existing physical hardware and technical ability is difficult to adapt to the demand of the smart grid, large data analysis and control, so based on the knowledge map design grid cloud data fusion method is fast. The first grid cloud data collection and processing, the second grid cloud data integration framework is designed, based on the knowledge map design grid cloud data fusion algorithm quickly, so as to realize the fast integration grid cloud data. The experimental results show that the proposed fast data fusion method has good fusion effect and has certain application value.

Hydrogen production technology is an effective means to solve the problem of energy shortage and environmental pollution at present, and is the basis for the development and application of hydrogen energy. This paper first analyzes and summarizes the current domestic and foreign hydrogen energy development policy planning, and discusses the current situation of electrolysis of water to hydrogen according to China's hydrogen energy development policy, and then systematically proposes the establishment of smart energy system park, the construction of renewable energy complementary hydrogen production system, and the construction of hydrogen refueling station hydrogen industry infrastructure scheme in view of the current problems in the electrolysis of water to hydrogen technology, and conducts a feasibility analysis of the scheme. A practical and feasible hydrogen refueling station construction and hydrogen-electricity complementary smart energy system based on electrolysis water hydrogen production technology have been determined.

In order to expand the application of the micro-pile foundation in transmission lines and reveal the uplift bearing capacity of micro-piles in mountain rock foundation. In-situ uplift load tests of micro-piles with different pile diameters and pile lengths were carried out. The load-displacement curve of micro-piles under load test and its ultimate anti-lifting bearing capacity were obtained. The uplift bearing characteristics of the single pile and group piles of micro-pile foundation in mountain rock foundation, and the variation of axial force and lateral resistance of piles were analyzed. The results show that the uplift bearing capacity of micro-piles is affected by the rock-socketed depth, and the bearing capacity is controlled by the strength of pile material when the rock-socketed depth exceeds a certain value. The variation of the pile diameter will affect the axial force of micro-piles and the distribution of pile side resistance. To give full play to the uplift bearing capacity of micro-piles, the reasonable pile diameter and pile length are needed. Each pile in the pile group foundation fails to give full play to its uplift bearing capacity. The uplift bearing capacity of the pile group is controlled by the tensile bearing capacity of the column.

The high curtailment rate caused by numerous factors is bound to affect the sustainable development of wind power, which is related to the realization, conduction, and appreciation of industrial value. Therefore, studying the influence factors of wind power and exploring their effects to improve the consumption capacity are crucial, which requires a more effective and efficient evaluation method. This paper proposes a novel evaluation model named interval type-2 fuzzy Petri net (IT2FPN) by combining interval type-2 fuzzy set (IT2FS) with fuzzy Petri net (FPN). The results are consistent with that in the reference, which proves the validity and applicability of proposed model.

External forces often cause mechanical damage to power system equipment, resulting in large-scale and long-term power outages. In this paper, the reliability of 220kV substations under external force is studied. First, a T-S fault tree that can be used to analyse the reliability of 220kV substations is established. And then, the fuzzy probability of failure of the fault tree basic event under three external force scenarios is obtained by the method of expert fuzzy judgment. Finally, the outage probability and reliability of the substation outgoing line under the three scenarios are calculated using the TS fault tree algorithm. This study is of great significance for further evaluating the reliability of the power system under external force.

Due to the dispersion effect of the traveling wave, the wave speed has frequency variation characteristics, and the error of the fault ranging method using empirical wave speed is large. To address this problem, this paper proposes a fault ranging method based on Multisynchrosqueezing Transform (MSST) for flexible DC transmission lines, the MSST method is used to perform time-frequency analysis of the fault line mode voltage traveling wave, to obtain the precise time and frequency information of the traveling wave arriving at the measurement points on both sides of the line, and to calculate the distance of the fault point according to the double-end ranging principle by combining the frequency variation characteristic curve of the wave speed. Since MSST squeezes the signal several times, it effectively improves the time-frequency aggregation of the signal, and the time-frequency feature extraction of the traveling wave and the identification of the wave head are more accurate. Finally, the effectiveness of the fault ranging algorithm was verified by the simulation of PSCAD and MATLAB.

Aiming at the problems of low monitoring frequency, low efficiency and poor accuracy of composite insulator temperature monitoring system for transmission line, a temperature monitoring method of high voltage AC composite insulator is proposed. The Maxwell equation is established by using the variational principle. Combined with the potential function, the electric field strength of the composite insulator of the transmission line is obtained, and then the electric field distribution of the composite insulator of the transmission line is calculated. The influence of the magnetic field force, temperature and arc resistance on the current in the arc time model is analyzed, and the arc speed control equation of the composite insulator of the transmission line is obtained. The arc time model of transmission line composite insulator is constructed. Combined with the temperature monitoring principle of transmission line composite insulator, the temperature of transmission line composite insulator is monitored. The results show that the proposed method can not only improve the temperature monitoring frequency and efficiency, but also improve the monitoring accuracy by reducing the temperature monitoring error.

This paper selects the PMSM (permanent magnet synchronous motor) controlled by the frequency converter as the research object, and simulates the distribution of the magnetic field and magnetic flux lines of motor in different states by using finite element software. It is found that due to the action of encoder, no matter whether the motor is stable at rated speed or not, the armature reaction always acts near the cross axis of the motor, and makes both ends of the PM have the effect of magnetization and demagnetization respectively. And the increase of load torque will also reduce the overall working point of the PM (permanent magnet). In the experimental part, two types of motors with partial demagnetization and uniform demagnetization were selected to carry out drag experiments and load experiments respectively, and the higher the degree of demagnetization, the higher the load coefficient, the more obvious the effect is.

Electricity spot market construction has become an important part of Chinese power system reform. Imbalance settlement risks have aroused the attention of relevant market participants. This paper analyzed the current power resource structure and market structure situation of Henan, gave examples on the influence of congestion fee by choosing different settlement methods, and risks caused by the market structure under the dual-track market situation in Henan. Mechanism designs on price, settlement, assessments are discussed to satisfy the need for development in the near future. Mechanisms including renewable portfolio standard, bidding, market clearing process, distributed energy resources are pointed out to meet the requirements of new energy system and carbon peak and carbon neutral goal in the standing future.

In order to meet the requirements of high operating life due to the frequent breaking of load current in the flexible DC power distribution system, and solve the problems of poor reliability of the fault self-clearing function with the converter and the high cost and large size adopting the DC circuit breaker scheme directly, this paper proposes a scheme of DC load switch based on coupled negative voltage principle. Firstly, its topology and working principle are introduced, and then the fast mechanical switch in the main branch based on double-acting structure and electromagnetic repulsion mechanism, the converter unit based on the coupled negative voltage device and the power electronic valve unit adopting anti-series topology, are designed respectively. Finally, according to the engineering requirements of the Pangdong central converter station in Wujiang district, Suzhou, a 10kV DC load switch prototype based on the coupled negative voltage principle is developed and verified by tests. The results show that the developed 10kV DC load switch can frequently break the rated current and the maximum breaking current within 3ms, the volume is the smallest with the same parameters in China, and it has the advantages of low loss, high mechanical life and high reliability. It fully satisfies the engineering requirements and provides a reference for the development of subsequent DC load switch products.

The access of distributed power sources has changed the structure and operation of distribution networks, which makes protection methods currently no longer applicable, and the small probability of extreme disasters makes protections of distribution network face new challenges. In order to make the protection of active distribution network response to different fault types at the same time, a new method of active distribution network pilot protection based on the comparison of the projected and actual values of voltage fault components is proposed based on the analysis of the voltage fault component characteristics of the distribution network when a fault occurs. The method can effectively identify the faults in the zone under different fault types, and has the advantages of not being affected by unmeasurable branches, not being affected by distributed power penetration, and having strong ability to withstand transition resistance.

With the continuous advancement of the electricity market, the demand for fine-grained subsidies and cost sharing of energy from different sources is getting higher and higher. There is an urgent need for a cost and subsidy distribution plan that is easy to use and can meet the needs of all parties. This paper proposes an intuitive, whole-process, and easy-to-use carbon emission and subsidy allocation method by using the proportional apportionment mechanism and iteration method, taking into account the energy loss of electrical devices and lines and the carbon emissions generated by the grid, as well as the various states of storages. We use an IEEE test feeder to implement our method and verify its effectiveness.

Compared with surface dielectric barrier discharge (SDBD), volume dielectric barrier discharge (VDBD) generates a wider range of plasma space, but the initial discharge voltage is higher, which is not conducive to the generation of discharge. Therefore, this paper proposes a dual-power excitation three-electrode DBD discharge device. On the basis of the traditional single-frequency excitation two-electrode volume DBD device, a third electrode powered by a high-frequency power supply is added to form a dual-power excitation three-electrode surface/ volume composite DBD device, which can reduce the initial discharge voltage of VDBD and improve active substance production. In this paper, the effects of the two discharge structures on the discharge characteristics and ozone generation characteristics of the device are mainly studied. The current graphs, discharge images, and power measurements show that compared with the single-frequency excitation two-electrode volume DBD, the dual-power excitation three-electrode composite DBD has more discharge channels, resulting in higher ozone concentrations. When the discharge gap is 2.8mm, the applied power frequency voltage is 16kV, and the high frequency voltage is 1kV, the volume discharge power and total discharge power of the three-electrode SVDBD excited by the dual-power excitation are 1.05 times and 1.16 times of that of the two-electrode VDBD excited by the single frequency excitation, respectively.

With the support of the increasingly mature material manufacturing process and the development of high-efficiency coal-fired power generation technology, countries have joined the research and development of 700℃ advanced ultra-supercritical technology, and made some achievements. However, in various reports and studies, it is found that countries have different opinions on the expected efficiency or coal consumption of 700 ℃ ultra-supercritical units, and there are great differences. This paper will summarize these differences, compare the effects of the current ultra- supercritical primary reheat and secondary reheat units in improving steam parameters to 700 ℃, and find out the problems existing in the heating process, it brings some new ideas to the development of 700 ℃ technology in the future.

In the high-power radio frequency ion source, a Faraday Shield (FS) with a thin-walled metal structure with slits is placed inside the ceramic tube to protect the wall of the RF ion source chamber from plasma erosion. Due to the complex structure of FS, it has a tremendous influence on the spatial distribution of electromagnetic field and produces a large amount of Ohmic loss in RF discharge. In this contribution, a self-consistent fluid model is applied to study the distribution of electromagnetic field and Ohmic losses on FS in the real plasma environment. The results show that the electric field in the plasma bulk exhibits a maximum at the slit gap, which makes the electrons gain more energy, resulting in a higher electron temperature. The peak of magnetic field appears near the inner side of the coil, which can be owed to the skin effect under high frequency. As for the Ohmic loss of FS, it is mainly distributed on the outer surface of the FS.

In the primary and secondary integrated switch, the distance between the primary equipment and the secondary equipment is very close. When transient processes such as switching operation and lightning impulse occur on the primary side, transient electromagnetic disturbance will occur on the secondary equipment, which will affect its normal operation. In this paper, a 75kV1.2/50us lightning impulse experiment of a 12kV primary and secondary integrated switch is carried out, and the induced voltage of the corresponding spatial electromagnetic field in a single-turn circular loop with a diameter of 15cm is measured. The results show that it is a damped oscillation wave with a peak value of 200V and a main frequency 8.1MHz. Through the test results of induction voltage, it is recommended that the shielding effectiveness of a shielding box enclosing the secondary equipment should be greater than 36.5dB, to ensure the normal operation of the secondary equipment. The shielding box is simulated, and the effects of operating frequency and hole size on shielding effectiveness for time-harmonic magnetic field are shown.

The power grid company has initially established costing plan of the main network substation equipment recently. However, the distribution network operation and maintenance cost allocation standard is extensive and the cost expenditure is affected by multiple factors, so that we cannot be simply allocated the cost according to the length of the distribution line. Through system data of the power grid company and field research by selecting typical areas, we sort out the factors that may affect the equipment assets of distribution network Operation and maintenance cost from external factors and internal factors and construct a cost index system, which mainly include environment, market, policy, equipment scale, equipment type, equipment quality, asset benefit and operation inspection strategy. Then, researchers utilize the panel data model to start the modeling analysis of the key influencing factors and identify the key influencing factors. Finally, taking the actual cost of Hubei Province as an example, the proposed method is verified, and the key influencing factors of the operation and maintenance cost of the distribution network are identified.

Compared with AC power supply system, DC power supply system reduces the current conversion link and improves the system efficiency, so it has great advantages to use DC power supply in buildings. At present, the domestic research on building DC power supply is still in the exploratory stage. The control strategy and coordinated control of converters in different operation modes are the keys to the stable operation of DC power supply system in buildings. In this paper, the control of the DC power supply system is divided into two layers: the lower converter layer and the upper management layer. The lower converter layer maintains the constant DC bus voltage and the upper energy management layer realizes the optimal and economic operation of the system. A multi-energy scheduling strategy of building DC power supply system based on time-of-use tariff and real-time battery control is proposed. The model is solved by using particle swarm algorithm to obtain the optimal operation of battery charging and discharging, buying and selling power from the grid according to PV output, load demand, battery status and time-of-use tariff, and the simulation proves the effectiveness of the multi-energy scheduling strategy.

In order to meet the challenges posed by the new power system to the operation and maintenance(O&M) business of EHV power grid and further improve the ability of multi-unit collaborative O&M in a wide area, according to the business execution characteristics of EHV power grid O&M enterprises, a construction architecture scheme of integrated production command and monitoring system based on cloud platform is proposed to support the integrated operation of relevant businesses.

Mobile devices get highly developed recently, but the energy supple of mobile devices still need to be optimized. This paper investigates wireless power transfer for mobile applications. The energy supplied by DC source is transferred through primary coil to the secondary coil using the principle of electromagnetic induction. A novel inductance structure is adopted at the primary coil to improve the quality factor of primary coil in a large range. The Q value of the primary coil used this novel structure can reach more than 600, and the overall efficiency can reach 75%. The maximum power transmission distance can reach 9cm at a resonant frequency of 104 KHz and more than one device can be charged for a multiple switch unit is implemented in primary subsystem. This research can make people charging more convenient, save the charging time when there are more than one mobile devices.

The electric power industry is the pillar industry of the society, shouldering the important task of providing stable power output for the country and the people. Therefore, the network security requirements of the power industry need to develop higher standards. The automation of transmission system has greatly improved the production efficiency of power grid, but the in-depth application of network technology has made the security of transmission system a new problem. In this paper, the network isolation technology is used to design the network isolation component. The Netmap high-speed packet receiving and sending framework technology is used to capture and forward the data of the network isolation component, reduce the number of copies of data packets in the kernel, and improve the packet forwarding speed of the network insulation component, which meets the needs of network isolation components in the case of large amounts of data. In the experimental test, according to the analysis of the test program and results, it is verified that the network isolation component can achieve access control and transmission protocol filtering functions in function, and can handle high concurrent connections in performance to ensure the efficient operation of the business, realize the isolation and exchange of internal and external network data of the transmission system, and protect the information security of the internal network of the system.

In this paper, combined with the construction of grid portraits, considering the differences between grids, and by sorting out the influencing factors of grid operation and maintenance costs, such as geographical location, operating environment, load density, economic level, safety and stability, a combined weight-based method is proposed. The cost optimization adjustment model corrects the grid operation and maintenance cost, realizes the fine evaluation and dynamic improvement of the operation and inspection standard cost level, and meets the needs of power grid operation management and cost refinement development under the new situation.

In the process of power transmission between regions, in order to expand the depth of power resource allocation, not only the power supply can be expanded in the region to meet the growing load demand, but also the transmission line can be expanded between regions. The construction of transmission line is conducive to peak regulation management, thereby reducing the investment in peak regulation power supply and the construction cost. In this paper, all kinds of index factors are used as input variables, and the system dynamics model is used to construct the regional energy supply and demand balance model considering transmission constraints. Firstly, the causality loop is constructed, and then the flow chart is used to construct the model. The variables in the system are assigned and defined. Finally, the coal supply and demand relationship, power supply and demand relationship and price trend among regions are predicted. The results show that the relationship between coal supply and demand will affect the fluctuation of coal price, thus affecting the economic development of each region. The economic development of both the sending end region and the receiving end region will increase the demand for power resources. The construction of power transmission channel can relieve the pressure of coal transportation channel to a certain extent, improve the absorption capacity of clean energy and promote the development of clean energy.

Low-power is a key technical indicator in the design of the comparator. This article discusses several different designs and techniques for achieving low power consumption. In this article, there are two different ways to achieve the goal. One is to transform the preamplifier of the comparator to realize low energy utilization. Such as using a tail capacitor with dynamic bias or adopting an inverter-based input pair. The other is to redesign the traditional comparator structure. For example, an edge-pursuit comparator and a high-performance dynamic comparator with a new latching stage. Due to the adjustment of different design parameters, the comparator also has advantages in terms of noise, delay, and gain.

This paper describes the development of the comparator in recent years. Some improvements were proposed based on conventional architecture, including the latch-type comparator with a dynamic bias pre-amplifier and the dynamic comparator based on a floating inversion amplifier (FIA). Some innovative architectures were presented to enhance the comparator performance, including the edge-pursuit comparator (EPC) and the edge race comparator (ERC). The techniques and achievements are summarized to facilitate the future development of the comparator.

This paper describes the development of the kickback noise reduction techniques used in comparators and the characteristics of the internal self-adjusting comparator. The Analog-to-Digital Converters are affected by the comparators most, which require further improvement of their basic parts such as the low offset voltage, high speed, and less resolution. With the use of the dynamic bias technique (DB), the comparator’s efficiency is improved. And the floating inverter peramplifier (FIA)technique reduce the increasing delay from the DB. To be the implement of the common-mode feedback (CMFB), an isolated voltage domain is utilized. The number of oscillation cycles (NOC) is put into the voltage-controlled oscillator (VOC) comparator, obtaining the delay correctly.

Current detection for cars flywheel more using the method of combination of contact type and non-contact detection steps is much and complicated, in this paper, the laser non-contact detection technology, using a laser displacement sensor for auto flywheel three parts of the circular runout error testing, flywheel model was constructed using the grid method and the laser displacement sensor based on the ant colony algorithm of path planning.

This paper introduces a spacecraft power supply applied to grid connected system. According to the grid connected power supply demand of spacecraft power supply, the three domain control strategy of satellite power supply is extended to five domain control strategy to realize the grid connected energy transmission function of spacecraft power supply. This paper introduces the principle of the five domain control strategy, gives the structure of the power supply, and finally tests the load step of the designed power supply.

Recently, along with the rapid development of the Internet and cloud computing, more and more network application protocols use encryption technology in order to protect users’ privacy. The detection of email high-risk operations such as cracking in encrypted traffic becomes a challenge. This paper proposes a DECinENT: a method to detect email crack in encrypted network traffic. The main idea is identifying mail behavior feature and detecting crack behavior by abnormal detection algorithm in the encrypted traffic. The DECinENT was used to test in experiment and compared with traditional methods. The results show that the efficiency has been greatly improved.

This paper presents the optimized memristor model based on the classical HP memristor model and the bipolar threshold behavior memristor model. This paper shows the mathematical modeling process of this optimized memristor model and the basic circuit principle of this optimized bipolar threshold behavior memristor model. The optimized bipolar threshold behavior memristor model is simulated by using OrCAD Pspice. We compared the optimized VTEAM memristor model with the classical HP memristor model in this paper and found the optimized bipolar threshold behavior memristor model has the superiority in switching characteristic and logical performance by the alternating voltage signal, which would be useful for new logical devices.

The flight control computer includes two parts, the power management system and the flight control computer system. The power management system realizes the input and output power supply management of airborne equipment, including components such as power management board, DC/DC power board, current acquisition board and power filter. The power module is converted into the +12V DC power supply required for the flight control computer to work， the current collection board collects the output current value of the relay components on it, and transmits it to the power management board. The current acquisition board collects the output current value of the relay components on it, and transmits it to the power management board. The power management board realizes AD acquisition of analog signals such as current and voltage, and the processing and analysis of voltage or current output signals such as fuel power supply, throttle steering gear, brake relay, inertial navigation positive pole, navigation light control, fuel level detection, etc. At the same time, the CAN interface is used to realize the communication with the flight control computer.

With the rapid growth of multimodal web data, the task of cross-modal retrieval, i.e., using a text query to search for images or vice versa, has attracted a lot of attention from researchers. Existing approaches usually learn a common representation space where different modalities can be directly compared. However, little work has been done to verify that the learned common representation space contains only common part shared between different modalities. In this paper, we present a coordinated and specific restricted Boltzmann machine (a.k.a. CSRBM) that can distinguish the common part from modality-specific part of different modalities. The proposed CSRBM consists of two RBMs, each with two hidden layers. The common hidden layer learns the common patterns shared within different modalities. And the modality-specific hidden layer learns the modality-specific patterns owned by individual modalities. To verify the split effectiveness of our proposed model, we construct a multimodal dataset based on the popular MNIST dataset. Moreover, we evaluate our model on three publicly real-world datasets with the task of cross-modal retrieval. The extensive experiments demonstrate the effectiveness of our CSRBM.

To evaluate the rehabilitation effect of elderly people during walking training on treadmills, a detection system capable of acquiring gait features was designed. Install the flexible force-sensitive sensor on the treadmill table, collect sensor data through STM32F205 microcontroller, and use W5300 Ethernet chip and STM32F205 microcontroller to build a TCP/IP network communication platform to ensure that sensor data can be synchronously transmitted to the host computer through the network port in real-time. The gait characteristics of the elderly are obtained by calculating the sensor data obtained by the host computer. System tests show that the gait characteristics obtained when walking on a treadmill are consistent with previous research results, thus verifying the accuracy of the detection system.

Deep neural networks perform well in many areas, but research has shown that they are vulnerable to adversarial example attacks. Therefore, adversarial attacks can be an important method to evaluate the robustness of the model before deep neural networks are deployed. However, the attack success rate of adversarial example in the black-box case still needs to be improved. There are numerous algorithms for attacking against neural networks, but most of the attack algorithms are slow, so fast generation of adversarial examples is gradually becoming a research focus in the field of adversarial examples. To address this situation, an adversarial example generation algorithm based on target detection is proposed. First, starting from the black-box scenario, target detection is performed on the original input image in the adversarial example generation, and constraints are added; then, through the improved differential evolution algorithm, the interference degree of each individual in the population to the image is calculated, and the individual with the best attack effect is continuously selected for iteration; finally, the adversarial example is generated based on the individual with the best attack effect, so as to eliminate the overfitting in the adversarial example generation and improve the adversarial This is used to eliminate the overfitting in the generation of the adversarial examples and improve the transferability of the adversarial examples. Experiments on the CIFAR-10 dataset validate the effectiveness of the proposed algorithm. The target detection-based adversarial example generation algorithm has a higher success rate of black-box attacks compared to the original one-pixel attacks.

With the development of science and technology, more and more artificial intelligence products are applied to the elderly, which is mainly composed of infrared communication function. However, there is no special device to detect the infrared communication function, so the quality of the artificial intelligence products can not be effective supervision. This paper studies and implements the infrared communication detection device, it mainly contains the six axis servo system, the detection module, the instrumented module and the power supply module. It can realize the infrared wavelength detection, the adaptive range detection of infrared receiving carrier frequency, the infrared light center frequency detection, infrared light interference detection and so on. The detection device this paper designs guarantees the quality of the artificial intelligence products which have the function of infrared communication in a certain extent.

With huge demand for electric energy and short storage of fossil fuels, renewable energy has been employed increasingly by many countries. However, transmitting this power into utility grid is still a big issue. In order to realize high-efficient power transmission and reduce power loss, various power converters have been developed to work in different transmission conditions. This paper presents topologies of 14 types of power converter with a typical PWM control strategy, including both DC to DC and DC to AC power converters with their simulation results, which are analyzed by MATLAB-Simulink. Based on the simulation, the performances are evaluated with different parameters, and advantages as well as disadvantages are concluded. Moreover, some advice on designing and choosing power converter under different transmission conditions are posted.

Elevator accidents are common. To ensure the safety of elevators, improve the efficiency of elevator maintenance and rescue personnel for the first time, this paper designs an elevator status acquisition terminal based on NB-IoT. STM32 is the core, and the elevator status signal is collected by installing sensors in the elevator car. After preprocessing, the collected elevator status information data is uploaded to the monitoring platform for monitoring through the NB-IoT module. In order not to be affected by the power failure, the system uses an uninterrupted power supply to ensure the normal collection of data.

This paper studies the automatic management technology of routing equipment. Using software defined network (SDN) technology, Python and virtual technology, this paper designs a batch configuration algorithm and a new router configuration algorithm which can configure the ports and protocols of the two devices at the same time when adds a router. On this basis, an automatic configuration software is designed and implemented. The configuration software includes remote login, information query, port configuration, routing protocol configuration, batch configuration of routing protocols, and automatic port and protocol configuration when adding routers, which solves remote login, configuration, and management problems of routing devices. Compared with other network automation configuration designs, the design of this paper has the following advantages: For the batch push of commands, this paper does not need prepare command documents in advance, and has good real-time interaction function; the automatically generated commands are pushed to the routing device with the information immediately available. When configuring two routers at the same time, you need to enter the login account and password, which is not automatically read from the password book. This ensures initial configuration security.

With the continuous cross-fertilization and development of automation control technology, computer technology, and information and communication technology, using control algorithms in the field of control to solve the failure of information and communication systems is a hot issue present. Based on the robust control algorithm, this paper designs a robust adaptive fault-tolerant control gain equalization equation, for the signal distortion fault problem caused by channel inter-code interference in the signal transmission of amorphous flattened air-ground wireless Ad-Hoc network nodes, and then solves the above fault problem so that the signal can be perfectly reproduced. The simulation results show that the signal distortion fault caused by inter-code interference is perfectly produced after fast adjustment under the control of the robust adaptive fault-tolerant control gain equalization equation designed in this paper, and the signal reproduction fluctuation error is within the allowed range. The signal reproduction error converges to zero asymptotically. This paper has significant research implications for subsequently laying down the economic framework for future low-altitude flight networking.

In order to better protect the environment, major cities have issued relevant garbage classification policies. At this stage, garbage classification mainly relies on manual sorting, which is cumbersome and unsafe. In order to efficiently classify and recycle garbage, use single-chip microcomputers and garbage bins. Combine it to create a trash bin capable of smart garbage sorting. The intelligent garbage sorting box uses STM32 single-chip microcomputer as the control center. It uses the garbage detection module, the motor transmission module, the full detection module of the garbage bin, the display module, the Zigbee information transmission and the upper computer module, which can quickly classify the garbage and remotely monitor it. A new type of garbage storage box. The intelligent sorting box can improve the efficiency of garbage sorting, simplify the garbage sorting process, realize the effective recycling of garbage, and improve the environmental quality.

To solve the problem of low accuracy of soft sensor of monitoring variables due to strong nonlinear, multivariable coupling, parameters time-varying and large time delays in complex chemical process, a soft sensor modeling method based on self-organizing interval type-2 fuzzy neural network (SOIT2FNN) is proposed. Firstly, to solve the problem that the structure of interval type 2 neural network (IT2FNN) is difficult to determine, an algorithm for self-organizing generation rules that uses firing strength and rule similarity to define growth and deletion indicators is proposed. The algorithm uses the firing strength to determine whether to generate rules, and deletes the rules according to the similarity, thereby determining the structure of the interval type-2 fuzzy neural network. Secondly, the relevant parameters of the SOIT2FNN model are corrected by the gradient descent algorithm. Finally, SOIT2FNN is used as a soft sensor model to detect the tail oxygen concentration for uncatalysed oxidation of cyclohexane process. The experimental results show that the soft sensor method based on SOIT2FNN model has the advantages of timely detection and high detection accuracy.

With the increasing demand for renewable energy, polycrystalline solar cells are becoming the mainstream of the solar cell market. Due to the significant impact of polysilicon surface texturing technology on improving the cell conversion efficiency of polysilicon solar cells, various surface texturing technologies have been developed by the scientific community. This paper is dedicated to a review of these technologies. Firstly, the technical principles of mechanical grooving, laser texturing, reactive ion etching, wet texturing and electrochemical etching are analyzed in turn. A more comprehensive review of these technologies is then presented in terms of quality, efficiency, cost and environmental friendliness. Finally, future trends in polysilicon surface texturing technology are predicted. The significance of this paper is that it provides a comprehensive review and summary of the various polysilicon surface texturing technologies.

This vehicle controller adopts a design based on the XC2287 processor as the core, integrates multiple communication interfaces such as serial port, LVDS, CAN, etc., and can receive input signals from voltage/current/resistance sensors, analog/switch input signals, and has PWM output signal, analog/switch output, etc. This controller is rich in types and numerous ports, making it suitable for a variety of different types of application requirements, and realizes the functions of sensors, switches and other mechanical transmission devices in the on-board control system of construction machinery. The controller's software and hardware adopt anti-interference design, and the shell is sealed and cast aluminum, which can meet the reliable application of construction machinery under harsh conditions such as wide temperature range, high humidity, severe vibration, and complex electromagnetic environment.

At present, smoke alarms have the disadvantages of not being able to send accurate fire positioning information in time and low reliability. In this study, a self-contained smoke alarm with dual-mode positioning function, high-precision smoke particle detection and rapid response is designed. The MC20 chip with integrated positioning and communication module is used to provide accurate positioning data and send alarm information in time.A composite smoke detection module with high-precision ion smoke sensor and photoelectric smoke sensor as the core is designed to improve the detection accuracy of smoke particles and reduce the rate of false alarm. Making a phone call and sending SMS at the same time are used as alarm method. The contents of the SMS are the positioning data and positioning link of the destination, and the link can directly jump to the terminal navigation map.

It is very important for wireless sensor networks to keep information transmission all the time. Because sensor nodes are powered by batteries, the network is easy to fail. Therefore, it is necessary to place relay nodes with high energy in the network to enable the network to transmit information. In practice, due to geographical location or other factors, not every location is suitable for placing relay nodes. In order to make the network more stable, the network should meet the requirements of multi-connectivity and multi-coverage as much as possible. Therefore, it is necessary to study the placement of relay nodes with position constraints that meet the k-coverage and S-connectivity. In this paper, the problem of relay node placement in a two-layer large scale wireless sensor network with position limitation is studied. Based on the convex hull algorithm, a relay node placement algorithm is proposed to meet K coverage and S connectivity, the algorithm complexity of O(n²) and approximate ratio of 1.5 are obtained.

In this paper, a novel ion-guided RF drive power is designed. The system consists of a control unit, a modulation part, a self-oscillation part and a negative feedback regulation part. Its simple structure, low cost and high integration promote the miniaturization of the mass spectrometer. The RF drive power only needs to provide a DC power supply, and it can rely on self-excited oscillation to generate RF signals to drive the multipole to realize ion transmission of the mass spectrometer. It is actually measured that the driving power is used for the hexapole ion transmission of mass spectrometers, the experimental results are good and stable.

The spark ablation solid injection method has the advantages of direct determination of conductive samples, low price, good repeatability and accuracy, and simple operation. However, the use of high-voltage DC power supply, ablative spark temperature is low, is not conducive to the formation of aerosols. In this paper, a high-energy spark pulse aerosol excitation source for the formation of aerosols for spark ablation metal samples is proposed, and the excitation source uses highenergy spark pulse discharge to continuously ablate metal samples to produce micro and nano-scale aerosols, which can be directly used for ICP mass spectrometry or spectroscopy for elemental analysis. The spark pulse discharge electric aerosol excitation source pulse energy of 855mJ, the ablation time is up to 125us, the ablation point is uniform and has a good remelting effect, and the particle size of the aerosol particles formed is uniform, which proves that it has high stability and reliability.

Photodetectors can capture optical signals and convert them into electrical signals. As an important optoelectronic device, it has important application value in imaging, optical communication, security inspection, environmental monitoring and other fields. Silicon-based photonic integrated chips have unique advantages and play an important role in optical communication. Fabrication of high-performance detector arrays is the current research hotspot and frontier direction. Driven by high-speed communications such as big data, cloud computing, and 5G communication networks, the transmission rate and carrying capacity of data information continue to climb to super-large capacity and high-speed. Data transmission is upgraded to larger bandwidth and faster rate, and electrical interconnection is difficult to meet the demand for high-speed data transmission. In this paper, different kinds of high-performance photodetectors are analyzed, and the performance characteristics of several typical high-performance photodetectors are studied. Optical communication technology has the advantages of abundant spectrum resources, high security, and strong anti-electromagnetic interference ability. It can carry long-distance, large-capacity and high-speed data communication, and has become an important development direction in data center interconnection solutions. Data center is an important carrier for processing and computing Internet applications. This paper studies the application of high-performance photodetectors in visible light communication technology in combination with data center communication scenarios.

The existing temperature compensating unit (TCU) of high-power circulator of EAST LHCD system has some disadvantages, such as unstable operation, high price, complex structure and difficult installation. These disadvantages will affect the performance of circulator in high-power mode, and then affect the stable operation of LHCD system. To solve these problems, this paper designs a new temperature compensating system (TCS). The new TCS takes STM32 as the main controller, and designs hardware circuits such as multi-channel temperature acquisition circuit, RS485 communication circuit and network communication circuit. At the same time, the upper computer monitoring program is developed to realize the functions of working state monitoring, real-time data display and operation data storage of physical device. The experimental results show that the new TCS can effectively maintain the performance of the circulator and provide guarantee for the high-power and stable operation of EAST LHCD system.

This paper builds on the academic background of learning to complete the production of solar cells and investigates the development of solar cells, their preparation processes, efficiency, and other aspects. This paper briefly introduces the development of solar cells, the preparation methods and conversion efficiency of solar cells in different eras and focuses on the future development of new materials to gradually improve the conversion rate of solar cells and find the limit of the conversion rate of solar cells.

The MIL-STD-1553B bus (1553B for short) is a centralized time-sharing serial bus. Among the current military data buses, the 1553B bus is the most mature and widely used data bus. This paper expounds a design method of 1553B bus communication system with CPCI bus interface. The host system communicates with 1533B interface module through CPCI bus interface. The core structure is FPGA(EP2C35F672I8N), CPCI bus interface bridge chip (PCI9054) and 1553B bus controller (61580), The design, debugging and testing of 1553B bus interface module are realized by using NIOS II soft core processor.

It is an effective measure for power enterprises to improve the voltage level of the system by raising and lowering the tap position of the main transformer. For the parallel operation of the main transformer, the whole process of adjusting the tap must strictly meet the conditions of parallel operation. Aiming at a case of backup protection action of main transformer caused by improper implementation of voltage regulation measures, this paper analyzes the uneven load distribution and increase of reactive power loss caused by improper adjustment of tap, and the analysis results are verified by PSCAD.

MXenes with unique two-dimensional (2D) open structure, high metallic conductivity, and tunable interlayer spacing are attractive for lithium-ion batteries (LIBs). However, the nanosheets restacking of MXenes hinders ion accessibility, resulting in the limited capacity, poor rate performance and sluggish ion-transport kinetics. Herein, phytic acid (PA), containing a myoinositol ring coupled with six phosphate groups, was first assembled into Ti3C2Tx MXene (M-PA) to achieve a supramolecular self-assembly via hydrogen bonds. The as-prepared 3D porous architecture of M-PA with larger interlayer spacing is favorable for the ion transfer, thus facilitating the improvement of lithium storage performance. As a result, the M-PA electrode exhibits a specific capacity of 391.7 mA h g^-1 at 0.1 A g^-1 (vs. 165.6 mA h g^-1 for the Ti3C2Tx counterpart) and outstanding ion-transport kinetics. This work highlights the delicate construction of high-performance MXene-based electrodes with 2D materials as the building blocks.

Facing the pressure of carbon emission reduction, thermal power units have basically achieved ultra-low emission of soot, but this has also brought greater water balance pressure to the desulfurization system, and the risk of fouling of the mist eliminators in the desulfurization tower has increased, affecting the operation safety of the unit. This paper systematically compares and selects the fouling treatment scheme of wet desulfurization mist eliminator from the aspects of demister type, structure and flue gas characteristics. A targeted air-water combined flushing scheme is proposed to provide technical support for the next field experiment.

At present, with the wide application of permanent magnet synchronous motor, the research on many performances of permanent magnet motor is getting deeper and deeper. In order to improve the back electromotive force waveform of permanent magnet synchronous motor and reduce the distortion rate of air gap magnetic field waveform. Parameterized design is used to optimize the shape of rotor punching shape and the width of stator slot. The motor model is established by Maxwell for simulation analysis. The analysis results show that the optimized magnetic density waveform of motor air gap is closer to sine wave, the torque ripple is reduced, and the servo performance of motor is improved.

In industrial production, the installation of speed sensors is not conducive to the miniaturization of motor control systems and will increase production costs. In order to overcome the above defects and realize the stable operation of PMSM at low speed, this paper studies a method of detecting the rotor position of PMSM based on high frequency pulse voltage injection. The high frequency pulse injection method relies on the salient pole effect of the motor, injecting high frequency pulse voltage into the stator winding of the motor, and then collecting the corresponding high frequency response current. The collected high frequency current signal is processed mathematically to obtain the position information of the rotor. The simulation results show that this method can ensure the stable operation of motor at low speed.

With the continuous improvement of urbanization, the cable rate of urban power grid continues to rise, the dense laying of high-voltage cables of various voltage levels in the same channel continues to increase, and the requirements for safe operation of cables are becoming higher and higher. The existing detection methods can not effectively find the problems of cables in the handover and operation stage in time, and there is a lack of on-site rapid detection and defect evaluation method, X-ray imaging detection technology has become the first choice for the internal defect detection method of XLPE insulated high-voltage cable. Because one exposure imaging can only obtain the gray defect information of one energy level in the detection process, it brings inconvenience to the cable detection work in high-intensity and complex environment. This paper proposes the application of X-ray intensity graded imaging detection method to cable flaw detection, the imaging quality and identification ability of X-ray detection of high-voltage cable defects are improved.

The application of SF₆ gas in electrical equipment is gradually limited due to its greenhouse effect on the environment. The suitable alternative gases are studied to replace SF₆. In this paper, the arc characteristics of SF₆, N₂, SF₆/N₂ mixture and dry air have been simulated in the ring main unit. The influence of different gases on the arcing process has been analysed.

With the advantages of simple control and controllable output energy and its direction, bi-directional DC/DC converter has an important application value in active equalization circuit of battery management system (BMS) in recent years. This paper in detail introduces a series of bi-directional DC/DC converter topologies, and the corresponding control principles, control methods and application in active equalization circuit.

The high-voltage side fuse of the generator outlet voltage transformer is used to protect the stability of the high-voltage system due to the failure of the voltage transformer itself or the failure of the primary side. When the fuse at the high-voltage side of the voltage transformer is overcurrent and blown, the voltage at the secondary side should drop to zero voltage instantaneously (Fast-melting). In actual operation, due to the aging, poor quality and partial fusing of the fuse, the quartz sand in it cannot play the arc extinguishing function well, resulting in the fuse breakpoint being continuously connected through the arc, and the voltage at the secondary side of the voltage transformer is sometimes high, sometimes low or continuously falling (Slow-melting), which leads to chain reactions such as false forced excitation, voltage rise at the machine terminal, alarm of the protection device, and machine tripping. This paper analyzes and studies the fusing mechanism of a slow-melting accident of the high-voltage side fuse of the generator outlet voltage transformer in a power station, and proposes targeted preventive measures.

The temperature rise measurement of transformer winding is one of the main items of transformer factory and preventive test. Most of the existing transformer winding temperature rise measuring devices are designed and measured according to the thermal resistance method specified in the transformer test standard. However, due to the inherent properties of power transformer, it takes a long time to measure the DC resistance of the transformer winding. Considering the particularity of transformer DC resistance measurement, the value of transformer winding inductance is estimated, and the resistance is inserted in series in the measurement circuit to reduce the time constant of the circuit, so as to achieve the purpose of fast measurement. Based on this method, an automatic measurement system for temperature rise detection is developed and the temperature rise is accurately measured.

Multi-core secondary cable in substation is large in quantity. Traditional cable detection method needs repeated measurement to determine the state of each cable core, which increases greatly the duplication of work. To resolve this problem, we propose a simple secondary cable fault detection method, the multi-core secondary cable is parallelly connected, and inject the DC pulse into it. The status of each cable core can be determined by the characteristics of reflected pulse. Reflected pulse contains fault characteristics of cables, which decays rapidly during propagation on other parallel cable. This method is verified by the experiment. The proposed method can detect typical faults accurately of multi-core secondary cable.

The flexible virtual inertial control of Direct Current (DC) microgrid can provide feasible solutions to the impact and challenges brought by a large number of distributed generators connected to the power grid. When the multi port converter adopts virtual capacitance control to further enhance the system inertia, it is necessary to study the cooperative allocation of Virtual Capacitance in the multi port converter. Therefore, this paper proposes a voltage cooperative control strategy of DC microgrid based on virtual capacitor control, so that each port can do its best and distribute cooperatively. Firstly, the total flexible virtual capacitance of the system is given. Secondly, the grey correlation analysis method is used to calculate the grey correlation coefficient of each end under the evaluation indexes such as state of charge and adjustable capacity of converter, so as to jointly distribute the total virtual capacitance of the system. Finally, the hardware in the loop simulation platform of DC microgrid is built to verify that the proposed control strategy can obtain better DC bus voltage response under power disturbance.

This paper establishes a three-dimensional numerical model of a 12 kV gas ring main unit in a C₄F₇N/CO₂ gas mixture during the interrupting process based on a magneto-hydrodynamic model, and uses FLUENT commercial software to simulate and analyze the arc motion and voltage characteristics of the existing 12 kV gas ring main unit interrupting structure in a C₄F₇N/CO₂ gas mixture during the interrupting process. By applying a magnetic field to the interrupter chamber, the effect of magnetic field regulation on the arc motion pattern and arc voltage is investigated, showing that the external magnetic field can make the arc enter the grid more and the arc voltage can be increased, and the arc voltage value will be larger just before the current crosses zero. At the same time, based on the synthetic loop platform, the arc-extinguishing structure of the ring main unit before and after the magnetic field adjustment is studied for its breaking capacity, which shows that the ring main unit can withstand a larger recovery voltage under the action of the external magnetic field and is more conducive to the successful breaking of the ring main unit.

This paper introduces a case study of an isolated neutral point system in which the overheating and melting at the flexible connection on the low-voltage side of unit auxiliary transformer leads to multiple intermittent grounding and subharmonic resonance, and finally an open circuit of the delta winding on the low-voltage side causes voltage waveform distortion, to study and analyze the case in all its stages based on the results of fault recording, KCL current equation, and Matlab simulation. At the end of the paper, it is proposed that for the transformer delta winding, the DC resistance between the connecting lines should be tested from the outgoing line terminals to eliminate the blind spot of the phase-to-phase DC resistance test, and at the same time, through the research and analysis, it provides a reference for the rapid inspection and analysis of hidden problems and hidden defects of similar equipment.

Nuclear magnetic resonance (NMR) is a nondestructive testing technology, which is mainly used in the aging state detection of composite insulators in the field of electrical engineering. To detect the aging state of composite insulator by NMR technology, the most important is to establish a static magnetic field with high magnetic induction intensity and high uniformity, which will help to greatly improve the signal-to-noise ratio of detection signal and obtain more accurate aging state information. At present, the establishment method of static magnetic field is mostly based on permanent magnet, but the static magnetic field with high magnetic induction intensity and high uniformity requires a large permanent magnet sensor, which does not have portability, so the detection scene is limited. In order to solve the problem of limited detection scene, a permanent magnet structure based on Halbach array is studied in this paper. The space volume is small and easy to carry. The magnetic induction intensity can reach 1.2T and the magnetic field uniformity can reach less than 10ppm.

The liquid metal current limiter based on the pinch effect has many advantages in the current limit, and has good application prospect in the power system. In this paper, a new LMCL which can string the cavity partition into the current limiting circuit is designed. The current limiting mechanism of the new LMCL is verified, and the arcing characteristics of the LMCL whose cavity partition are insulating material and conductive material are compared by experimental research. The experimental results show that the cavity partition material with higher electrical conductivity can effectively reduce the intensity of arc combustion, and the cavity partition can be connected to the circuit to limit the current, which provides an important reference for the optimal parameter design of the LMCL.

In order to ensure the safe and reliable operation of offshore wind farm, this paper analyzes the configuration scheme and operation mode of standby diesel generator set under the condition of isolated networks of offshore wind farm. Considering the diesel generator capacity accounting, reactive over-voltage, short-circuit current, harmonic analysis and protection setting under the condition of isolated grids of offshore wind farms, the rationality of the standby diesel generator set scheme was verified. The results show that the proposed method is real and effective.

This paper analyzes the influence scope, emergency disposal, fault causes and risk characteristics of Jilin snow disaster, and discusses the means of emergency disposal. The paper summarizes the operation years and icing history of optical cables in Henan Province, excavates the causes of icing, analyzes and studies the prone to icing in west, north and central Henan from the perspective of geographical environment, analyzes the relationship between icing and optical cable aging from the perspective of operation years, and focuses on preventing nine optical cables in the above areas. Combined with the weather change, the emergency detour scheme of icing optical cable is revised in a rolling way, and come to the conclusion of using multiplex protection to facilitate the emergency detour of channels, and increasing OTN coverage rate to facilitate the emergency detour of optical path, so as to cope with the risk of icing optical cable during winter peaks.

Accurate and efficient video quality assessment (VQA) methods are important guidelines for optimizing network video quality, improving video compression performance, and recommending compression coding parameters. No-reference VQA methods are mainly divided into bitstream-based methods, which are less time-consuming but less accurate, and pixel-based methods, which are more accurate but more time-consuming and require high hardware requirements. Both methods are not applicable to the quality assessment of a large number of network videos. To address the above problem, we propose an efficient and accurate cross-domain no-reference network video quality assessment method (CDNRVQA) based on bitstream. CDNRVQA takes I-frames as representative frames and uses deep neural network to obtain video high semantic features and distortion features from the representative frames, then obtains temporal features and some spatial features of the video based on the macroblock and motion vector in the compressed videos, and uses feature fusion strategy to fuse features from pixel domain and compressed domain, and finally obtains comprehensive cross-domain features which can represent the video distortion brought by content, motion and compression. CDNRVQA performs well on large VQA datasets. CDNRVQA achieves similar performance to the state-of-the-art VQA models, but greatly improves prediction time, and greatly reduces hardware consumption on the device. CDNRVQA makes it possible to apply VQA methods to video platforms.

While dealing with transient target absence or sudden position change, short-term visual object tracking is not capable of continuously spotting the target. Therefore, long-term tracking methods are indispensable for handling these issues. It is necessary to summarize these long-term tracking methods. In this work, we propose to demonstrate some long-term tracking methods including LTMU, ST Transformer and GlobalTrack and three benchmarks (OxUvA, TLP, and CDTB), and we aim to compare these methods based on various benchmarks. All those methods achieve distinguishing performance on those datasets. Furthermore, we are going to present some prospects of the long-term tracking.

Automatic observation for calibration of satellite remote sensors is used widely in the past few years, which usually two or three automatic observation instruments for necessary characteristics, such as surface reflectance, aerosol optical depth and diffuse irradiance for more calibration accurate, are placed in the site. In this paper, an automated precision ground-solar-sky radiometer (GSSR) is developed with constant temperature for Si and InGaAs detectors avoiding the influence of temperature drift of detector responsivity and BeiDou terminal for data transmission. The radiometer is equipped with three aperture tubes, one for measuring the direct solar irradiance and the sky diffuse radiance, and the other two for measuring the surface reflectance. When the satellite overpass, it can adjust to consistent observation azimuth with the remote sensors eliminating the effect of bidirectional reflectance. The radiometric calibration of irradiance channels are carried out with Langley method and the radiance channels with the lamp-panel system (LPS) method and the solar radiation-based calibration (SRBC) method. The deviation of the calibration coefficients obtained by two independent calibration methods LPS and SRBC is less than 3% in band 1-7, indicating that the calibration results are accurate.

Path planning and navigation and positioning are the key issues in the operation of the book return smart car. The book return smart car system proposed in this paper adopts the improved A^* algorithm and inertial navigation for path planning and navigation and positioning. First, use the camera to identify the map of the library. Secondly, apply the A^* algorithm to solve the optimal path from the initial position to the final position. Finally, the icm20602 six-axis IMU sensor integrates the acceleration during the driving process of the book-returning smart car, and the encoder is used to locate the vehicle's position.

5G (5th Generation Mobile Networks) communication, with its large bandwidth, massive connection, high reliability and low latency, as well as its flexible deployment mode independent of the grid, can provide the key technology to support the transformation of the distribution network towards digital grid and improve transparency as well as intelligence. However, the terminals do not have strong access capabilities in the practice engineering application, and their cooperation with the power equipment are not close as designed. In this regard, this paper first summarized the signal attenuation influencing factors and transmission effect enhancement methods in the propagation process of 5G air interface, elaborated the deployment scheme of multi-access edge computing (MEC), further sorted out the ideas of MEC and power facilities integration deployment, which provides some reference for the application of 5G communication in distribution engineering.

Taijiquan is an important Chinese Kung Fu in the study of movement balance. This paper discusses the system solution of motion stability of humanoid robot from the perspective of motion stability of Taijiquan itself. This paper starts with the systematic practice theory of Chinese Taijiquan in human dynamic balance. Combined with the correlation between humanoid robot structure and human body structure. The spiral balance and wire winding force system are used to deal with the problems of self stability and motion or load stability of humanoid robot, and then the system design scheme to realize the motion stability of humanoid robot is proposed. The author hopes to provide a novel research perspective for the current humanoid robot dynamic balance from the perspective of inheriting the human body dynamic balance theory for hundreds of years.

Ski jumping is a fast and wide-range motion. Although wearable devices are available to analyze the motion, it is cumbersome and difficult to implement. Since video data is relatively simple to obtain, this paper proposes a video-based method for estimating the pose of ski jumpers. In this method, we use a high-speed camera as a video data collector, and use Simi Motion software to convert the video into frames and manually annotate keypoints. The video data of three athletes is used to build the training set, and another is used to build the test set. In addition, we use High-Resolution Net (HRNet) to transfer the learning of feature knowledge from the public dataset COCO2017 to the task of ski jumpers pose estimation. The experiments show that under the real labeled bounding box, an average precision of 84.6% is obtained, which is higher than other mainstream human pose estimate methods.

Unsupervised person re-identification (re-ID) aims to identify the same persons' images across different cameras by training on unlabeled data. In which, how to alleviating the occlusion problem in unsupervised person re-ID is a great challenge. Recently, the work on unsupervised re-ID has achieved substantial progress by clustering on the unlabelled target data or unsupervised domain adaptation. Nevertheless, previous re-ID methods either ignored the occlusion problem or solved it based on extreme assumptions. Therefore, in order to develop a kind of more practical and generalized re-ID methods, this paper propose to alleviate the occlusion problem for the unsupervised model. Firstly, we introduce a poseguided branch to extract the key-points information of person. Then, the global feature extracted by backbone and postural feature obtained by pose-guided branch are fused to fed into the unsupervised system. Finally, the experimental results demonstrate that our identification accuracy has achieved strong performance on the three person re-ID dataset.

In the wake of COP26 climate summit in Glasgow, countries around world began to tighten their regulations on carbon emission and prioritize research targeting energy conservation. One of the fields that energy conservation can be promoted is oil exploitation, where pumping units are the main facilities. The pumping units on oil fields are critical for energy production around the world, they often operate in extremely tough and complex environment and have a range of parameters to be managed and controlled. In the operation process of pumping units, there is great quantity of potential energy wasted. Aiming at addressing these issues, a solution of oil field pumping unit energy recovery management system was proposed in this paper, which can realize functions including monitoring and control of the pumping units, and management of the energy storage modules. The energy recovery management system was programmed under C# programming language and MVC architecture, and was improved using the SQLite database. To optimize the outer layer, the User Interface was enhanced with DevExpress toolkit. Apart from that, measures were taken to ensure the security and stability of the system. After systematical design, software test was successfully carried out, which proved this design is safe and stable and can effectively serve as energy recovery management system.

In order to better solve the social hot issue of garbage classification difficulties, we designed an intelligent garbage classification system based on the lightweight convolutional neural network MobileNet V2. The system can automatically realize the intelligent classification operation of garbage, and liberate people from the tedious and mechanical garbage classification. The system uses the Pytorch deep learning framework to build the MobileNet V2 network and train the garbage classification model, so that the model can be mounted on the Raspberry Pi to achieve embedded integration. This design realizes the highly integrated, intelligent, and data visualization of the garbage classification system, with accurate classification It has the characteristics of high density, simple and convenient use, high practical value and good applicability, which can greatly simplify the process of garbage classification.

With the development of distribution network, load importance is sometimes different from load demand, the load shedding strategy made during fault recovery often results in the reduction of user satisfaction. Therefore, this paper proposes the idea of introducing non-cooperative game into the distribution system fault recovery model, and regards the fault recovery problem as a two-players game between grid and user. Firstly, the improved adaptive genetic algorithm is used to solve the fault recovery model to obtain the initial recovery strategy. Then, the user satisfaction is calculated and used as the condition for the user to initiate the game to the grid, then the recovery strategy is dynamically adjusted. Moreover, all the strategies that satisfy the conditions are selected to construct the recovery strategy set. Finally, the Nash equilibrium solution which balanced the participants’ interests is solved from the strategy set according to the game criteria. The proposed strategy which finds the optimal recovery strategy that meets the minimum total payment value and the highest user satisfaction is verified based on the modified IEEE33 node model.

With the development of the Internet, many log data are generated in the server. How to ensure the security of log data in the server is a problem that needs to be solved. At present, log data storage in most servers relies on a centralized log database or local files. With the increase of running time and scale, there are great challenges in data security and stability. Meanwhile, attackers can tamper with the server's logs to evade tracking. To solve the above problems, this paper proposes a secure multi-level log data storage system based on permissioned blockchain and permission-less blockchain, with high throughput and low cost. In addition, different encryption methods are adopted for different levels of log data to provide higher security

In view of the fact that the existing electric vehicle (EV) charging load model does not consider the fuzziness of EV return time, the random fuzzy theory is introduced into the EV charging load model. Based on the uncertainty theory, the return time of electric vehicles is defined as a random fuzzy variable, and an electric vehicle charging load model including random fuzzy is established. Using the smart grid, the orderly charging strategy of electric vehicles is solved and optimized. Finally, the simulation example is analyzed, and it is confirmed that the electric vehicle charging load model with random fuzzy is more practical than the original model, and the electric vehicle charging strategy can achieve orderly charging effect.

In modern interior design, there have been many problems with the quality of interior design schemes. The aim of this study is to improve the quality of interior design solutions and to ensure the standard of interior decoration design. Based on the existing evaluation standards for interior decoration design solutions, we make full use of artificial intelligence science and technology to analyse various indoor environmental quality indicators, establish an evaluation model for interior decoration design solutions, make the design more human-centred and effectively meet the needs of users, and establish an evaluation system for interior decoration design solutions under artificial intelligence processing technology through experimental verification analysis. This is both a need for interior design marketing and an inevitable trend in the future development of interior space design solutions.

Organic solar cells (OSCs) have received much attention due to its low cost and lightweight. Many researchers have been down to improve its efficiency, which has recently increased to about 20%. However, the stability of OSCs is still hard to control since different organic materials present various properties. Therefore, the development of stability is one of the most significant directions to promote the commercialization of OSCs. In this paper, recent research on the stability of OSCs is reviewed to introduce the mechanism of degradation. Based on the cause of degradation, the strategies to improve the stability are concluded from the perspective of humidity, temperature and light.

The Internet of Things allows everything to be interconnected, but there is also the risk of malicious devices or vulnerable devices causing damage to the network. In order to avoid this risk, how to accurately identify which devices are connected to the network in a resource-limited IoT environment becomes an urgent problem to be solved. Identification by extracting network traffic features as device fingerprints is a more effective and low-interference method today. This paper proposes a device identification method based on device traffic fingerprints, which utilizes convolutional neural networks (CNN) to extract features from device network traffic, and then uses SVM for device classification. The experimental results show that, compared with the CNN method and the method of manually extracting features, the identification accuracy and stability of the method proposed in this paper are significantly improved.

This paper aims to solve the disadvantages of manually checking the thesis format and adjusting many formats. This paper designs an automatic format detection method and transform method. The automatic format detection method can adapt to papers with different structures and formats by configuring the relevant information required by different formats. The automatic format transform method converts the main elements of the thesis through the configuration of the relevant information required by different formats. The experimental results show that the automatic format detection method and the automatic format transform method are suitable.

The photovoltaic power system is the most rapidly developing and promising renewable energy industries current. And the market nowadays is dominated by silicon-based solar cells, while the structure and the material for the more advanced third-generation solar cell have been proposed. For the different designs, the application is varied. Like for satellite, the solar panel is mainly made from GaAs, while the photovoltaic on the ground is silicon- based due to the cost for generation, producing efficiencies, mature producing technologies, and supporting industries. The aim of this work is to introduce the three types of third-generation solar cells and evaluate their latest electrical performance. On the other hand, the field for the application is presented, which can be helpful for the analysis of applications for different solar cells. A concise table for comparison is provided at the end to summarize their performance concerning the applications fields.

The identification of argument relation is an important subtask of argumentation mining. Its purpose is to identify the support or attack relationship between two argument components, so that people can understand the argument process in the argumentative text more deeply. This paper proposes a research method based on Scibert and BILSTM-CRF model. First, the pre-trained language model Scibert dynamically obtains word vectors, and then combines with the BiLSTM network to fine-tune downstream tasks to obtain contextual information. Finally, the argument relation is identified by conditional random field. Experiments were conducted on two argumentation mining datasets, Persuasive Essays and UKP Sentential Corpus, which were published in Germany. The experimental results show that our method is better than the baseline method.

A reconfigurable robot based on FPGA is designed and implemented in this project, which takes DE10_nano development board as the control core, integrates MPU-6500 gyroscope, ultrasonic module, photosensitive sensor, human body recognition module, and Openmv module, and expands the functions of obstacle avoidance, following, automatic light source tracking, human body recognition and so on. After many times of debugging, the automatic stabilization time of the robot is less than 2s after power on, and the obstacle avoidance function can be realized by automatically detecting objects in 10cm, recognizing the color of objects, and following them within the range of 20~30cm, automatically identifying light sources and tracking within 2m, and identifying the human body and alarm within 6m.

In response to environmental changes, the software should have dynamic response capabilities. To this end, this paper proposes an intelligent software framework that can be evolved, reasoned, and dynamically programmed. The framework of this paper is based on SpringBoot, combined with the use of hot deployment mechanism to realize the framework's evolvability. The frame reasoning part is implemented based on the Rete algorithm. Based on the written rules, the reasoning engine can infer the next state of the system based on the environmental data monitored by the system, and use the state information to complete the equipment control. The ability of Framework programming is reflected in two aspects: the replace-ability of the framework code and the reprogram ability of the rule code. The research results show that the framework has good adaptability in dealing with environmental changes, system changes, and rule changes.

The distribution network construction has the characteristics of many points, and with the gradual improvement of network construction, distribution grid investment management focus, therefore, scientific and reasonable construction of distribution grid input-output evaluation index system and evaluation method, can effectively improve the science of the grid enterprise distribution network investment decision-making, and improve the level of business efficiency. Based on the characteristics of the distribution network construction and the future development direction, we construct the evaluation index system and evaluation method based on DEA theory, and verify the effectiveness of the method model through empirical analysis. This method and model can effectively guide power grid enterprises to improve the scientificity of distribution network asset investment, and improve the efficiency and efficiency level of enterprise distribution network investment.

In order to improve the health monitoring effect for patients with chronic diseases, and make it more comprehensive and efficient, this paper proposes the design and implementation of a smart medical system that can prevent and monitor chronic diseases. The system has friendly interface and simple operation. Through big data technology, it can monitor patients' physical signs data in real time and make correlation analysis of patients' medical history and historical physical signs data, so as to effectively and timely assess and intelligent early warning of patients' health.

In order to improve the efficiency of on-site management of UHV projects. This paper discusses the application method of the Internet of Things technology in the intelligent management of the UHV project site. The method uses the mobile phone UHV project information of the Internet of Things technology. Based on the collected information, it builds an on­site information management and evaluation system for UHV projects. So far, this paper has realized the research goal of intelligent management ofUHV project site. Finally, experiments show that the application method oflnternet of things technology in on-site intelligent management of UHV project has high practicability and fully meets the research requirements.

UAV has become a promising development direction in 5G era because of its flexible deployment and economic efficiency. UAV with communication function can serve many scenes, such as traffic congestion, limited base station conditions, emergency rescue and so on. However, UAV has limited airborne energy, throughput and energy efficiency are the main bottlenecks of UAV as an air base station. Based on the consideration of various factors such as channel, user, UAV speed and transmission power, this paper constructs a reinforcement learning model for UAV energy efficiency, and puts forward the description of environment matrix to quantify the environmental parameters and participate in the action value evaluation. Firstly, based on the existing conditions, a constrained model is established to maximize the information throughput per unit energy consumption by combining historical empirical data with the exploration of a certain degree of freedom. In addition, we establish strong constraints on UAV energy to avoid unnecessary consumption as much as possible. The experimental results show that the algorithm proposed in this paper shows good performance in the simulation stage and excellent stability in the open environment.

The application scenarios of face recognition are very wide, and it has become a research hotspot in the field of computer vision. The deep learning technology represented by the convolutional neural network has shown great advantages in the field of face recognition. Its network structure is developing in a deeper and wider direction, and has achieved better and better recognition results. Due to its complex structure, the training is difficult to converge and occupies a large amount of memory and takes a long time to load, which is not conducive to the application in practical engineering. In order to solve this problem, this paper studies the application of deep learning in the field of face recognition, and proposes an improved residual neural network. On the basis of the original model, this model sacrifices less accuracy, and improves the speed of face recognition to a greater extent. This makes the model of great significance in the field of engineering practice.

With the rapid development of the electricity market, the idea of improving the flexibility of the power system has gradually shifted to the load side. The load side is regulated by means of Demand Response (DR) so as to reduce the regulation pressure of the power supply side, change the situation that the traditional grid power supply passively follows the load change, and realize the two-way response of the source and load. Load is no longer a simple deterministic model but a probabilistic model. The power balance relationship will also be transformed from one-way type to source-load complementary type. In this paper, the effect of load side flexibility regulation means on the system optimization configuration of power supply side regulation capability is studied. The electricity price type and incentive demand side response are applied to the power grid to study the intelligent regulation strategy of the power grid to obtain the improvement of the flexibility performance of the power grid. On this basis, in view of the call priority of all kinds of flexibility resources, the adjustment ability and capacity ratio of all kinds of flexibility resources on both sides of the source and load are comprehensively considered to complete the priority division of flexibility resources on both sides of the source and load.

IoT applications consist of a series of correlated IoT services. Compared with traditional services, IoT services have new features in distribution, dependency, security and dynamism, which pose new challenges to the assurance of performance for cloud-based IoT services. Service scheduling on the cloud is an effective approach to guarantee service performance. Taking into account the complex contextual information of IoT services, this paper proposes a Context-Aware Online IoT Service Scheduling Approach in Cloud Environments. First, based on the characteristics of IoT services, a description of the IoT service scheduling problem is given, including the constraints of transforming the contexts of IoT services into scheduling. Then, the service scheduling problem is mapped to a scheduling plan prediction problem based on a neural network approach. To verify the effectiveness of the method, two baseline algorithms were selected and simulated on the WorkflowSim platform together with the method proposed in this paper. In comparison, our method is found to have similar scheduling costs with the optimization method, but can improve the efficiency when generating the scheduling plan, which can better meet the demands of low time delay for online scheduling under the conditions of dynamic changes in IoT context.

We present a target discovery simulation method and software, including target environment module, photoelectric system module and search simulation module. The target environment module includes target / background radiation characteristic parameter module and transmission module, and the photoelectric system module includes visible light imaging system and infrared thermal imaging system, the search simulation module includes search model and dynamic detection probability model. This method can simulate the process of visible and infrared target discovery.

This paper designs a smart car that can automatically deliver meals in dormitories, isolated hotels and other scenarios. This system uses i.MX RT1064 as the main controller, and completes the route tracking and room number recognition of the smart car through the MT9V034 camera and the OpenART mini visual sensor module respectively. The target detection method is the SSD algorithm in the one-stage method. After optimization, the recognition rate is as high as 90%, which can successfully complete the meal delivery task. This system greatly reduces the risk of human-to-human contact, reduces the probability of contracting COVID-19, and contributes to epidemic prevention and control measures to minimize risks.

As an important enabling way to realize digital transformation and promote intelligent upgrading, digital twins have been attracting attention from all walks of life, and have moved from theoretical research to practical application. At present, most domestic elevators still use the traditional dispatching system. The traditional elevator dispatching system can't realize intelligent elevator man-machine interaction, which leads to a series of problems such as low efficiency and poor linkage of elevator group control. Thus, this paper designs an intelligent elevator dispatching system based on digital twins, which combines digital twins with artificial intelligence identification technology, internet of things, cloud database technology and so on. The system consists of four parts: physical elevator model, virtual elevator model, elevator twin data model and mobile phone client. Furthermore, the rational allocation of elevator resources is strengthened, and the interaction ability between elevators and passengers is enhanced. While improving the comfort of passengers, it also reduces the energy consumption of the elevator macroscopically.

Knowledge extraction is a key task in the construction of essay knowledge graphs. This paper summarizes and analyzes the progress of research related to element and relation extraction, using discourse structure as the knowledge level. We find that existing methods treat element and relation extraction as two independent tasks, both relying on contextual information, without considering the semantic interaction between the two tasks. Thus, we propose the joint extraction of elements and relations as a novel task, migrate the joint method of entity-relation extraction task to this task, and focus on the research advances of the joint method to summarize.

The study of hand gesture recognition has significant academic value and wide application prospects, but the accuracy of hand gesture recognition is affected by differences in environment, skin color, and hand shape. In this paper, we first preprocess the dataset, obtain the key points of the hand through the MediaPipe framework, and draw these key points on white pictures to form a new dataset, and then train the unpreprocessed dataset and the preprocessed dataset separately using the ResNet50 network structure. The results of the comparison experiments show that the accuracy obtained from the preprocessed dataset is 1.9% higher than that of the unpreprocessed dataset.

Under the goal of "Carbon Peak, Carbon Neutrality", clean energy generation will gradually become the main part of power supply. In order to stabilize the output fluctuation of wind and photovoltaic power generation, and improve the efficiency of clean energy generation and reliability of power grid, this paper designs a multi-energy complementary power generation system with pumped storage power station, and analyzes the system architecture, operation mode, construction mode. A pilot system is built to analyze the 24-hour operation process of the system under four operation modes. The single day benefit of power generation are quantitatively evaluated, and the benefits of ecological civilization construction and stable operation of power grid are expounded, so as to provide a reference for the largescale development and utilization of clean energy sources.

The maintenance of power transmission lines is an important part in avoiding potential safety hazards and protecting residents’ electricity use. Currently the inspection of power transmission poles mainly relies on manual positioning, which is not efficient. To improve the efficiency in the positioning of poles, this research proposes a real-time detection method for power transmission poles based on video streaming media and AI algorithms. The method obtains the real-time status of roadsides by sending video streams through the onboard camera and receiving streams through OpenCV and utilizes YOLO, DeepSORT and a fault-tolerant algorithm to detect, track and count power transmission poles. The result of test data shows that the proposed method could not only detect the poles along roadside with good precision but also count the poles accurately with a success rate of over 90%, which could greatly assist in the inspection of power transmission poles.

Urban distribution networks (DNs) are facing challenges in keeping power service for lifeline loads under extreme events such as storms. However, the economy under normal operation of resilience-oriented measures need to be further considered in resilience enhancement strategies. Therefore, this paper proposes an equilibrium configuration method for mobile energy storage systems (MESSs) in order to improve both the resilience and economy of urban distribution networks against storms based on Nash bargaining. Considering the accurate failure rate model of urban DN components, an equilibrium configuration method for sizing MESSs is proposed, in which the Nash bargaining method is used to deal with the paradox between economy and resilience. Simulation tests further verify the superiority of the proposed equilibrium configuration strategy of MESSs considering multiple objectives in urban DNs.

With the aggravation of global energy shortage, it is imperative for new energy to be connected in a high proportion in the power system. In the context of a new power system, the high proportion of wind power access brings greater challenges to the recovery process after a major power failure. Therefore, it is of great significance to study wind power uncertainty processing methods for evaluating the impact of wind power black start value. Due to the correlation between multiple wind farms, Copula theory was first used to analyze, and Latin hypercube sampling was used to get the typical wind speed scenarios with correlation, and then the synchronous back-substitution method was used to subdue the typical scenarios. On this basis, the fuzzy set based on KL divergence was established to obtain the wind power output sequence. Then, the risk value theory was introduced to establish the wind power black start value model considering the extreme output of wind power. Finally, IEEE 39-bus system with wind farms was given to verify the effectiveness of the proposed method. The applicability of wind power black-start value is improved.

In order to quickly and accurately identify the frontier research hot spots in the subject field and provide accurateprediction service for scientific researchers, this paper takes the scientific and technological literature in the fieldof coldchain logistics as the research object, and proposes a BTG (Bert-Based Topic Generator) model, which extracts thesubject words from the titles, keywords and abstracts of the literature. In this model, the pre-trained language model Bert is used to obtain the word embedding, which is dimensionally reduced by the UMAP algorithm, HDBSCANclusteringalgorithm is used for clustering, and TF-IDF weight is calculated into the core subject words in each clustering category. This study conducts comparative experiments on HDBSCAN, K-means and DBSCAN clustering algorithms byusingevaluation criteriones such as degree of cohesion and degree of separation. According to the experimental results, theDBSCAN algorithm outperform other clustering algorithms based on Separation criterion, Intra-Cluster part ofDispersion, Inter-Cluster part of Dispersion and PR criterion.

As the most widely used clustering algorithm nowadays, the K-means algorithm has been applied in diverse fields. It is characterized with a fast calculation speed and a simple algorithm. It is, however, nonetheless beset by a number of problems. One difficulty is that choosing the starting center at random will have a negative influence on the clustering result. Second, outliers are susceptible to the K-means algorithm. Third, such an algorithm still has a significant time cost. To address these concerns, this work introduces the AGK Adaptive-GK method, which takes use of the grid clustering technique's benefits. Our AGK can accurately identify an initial center, increase the accuracy of the standard K-means method, and lower the algorithm's computation complexity. We did a thorough study of our AGK on a variety of data sets, and the findings show that it is accurate and efficient.

This paper combines fuzzy inference algorithm with the simulated annealing modified Q (λ) learning algorithm to solve the problem of robot who avoids obstacles in unknown environments. The system learns autonomously by itself without supervision or any prior training data. One of the popular methods for path planning algorithm is improved simulated annealing Q (λ)-learning. The fuzzy logic control is utilized to solve the generalization of continuous space. The combination of the two algorithms can solve the generalization of the continuous state space. Another advantage is that it can reach the balance of the exploration and utilization of the action strategy. The Simulation results exhibit the effectiveness of the proposed method compared with the original method.

Data cleaning which refers to all kinds of activities to detect and repair errors in the data, is one of the most important problems in data management. This paper proposes a multi-rule data cleaning tool that combines multiple dependencies for error detection and error repair. As an example, Conditional Functional Dependencies (CFDs) and Related Dependencies (RDs) are combined in experiments to detect violations and repair erroneous records in data, which shows the validity of multi-rule method.

The use of scheduling algorithms is the key to bandwidth management in the field of network transmission. The quality of service guarantee of data streams needs to be combined with proper queue scheduling algorithms for the network QoS to be maximized. Fairness, delay performance and bandwidth utilization are the criteria for evaluating queue scheduling algorithms. The fairness and delay of the algorithm are the main reference indicators for the design of the scheduling algorithm, and the simplicity and scalability of the algorithm are guaranteed as much as possible based on the above two indicators, but the complexity of the algorithm is often contradictory to the delay and fairness, so it should be considered in an integrated manner to achieve the optimal design. In this paper, we comprehensively review the literature on this topic and propose classification criteria to present and qualitatively compare existing queue scheduling algorithms based on the above algorithm evaluation metrics.

The back EMF method is usually used in the operation control of sensorless BLDC motor, but in the starting stage of the motor, because its speed is too low, it is difficult to accurately detect the back EMF, resulting in starting failure or out of step reverse rotation of the motor. Aiming at the starting problem of BLDC motor without position sensor, this paper proposes a new starting algorithm. Compared with the traditional three-stage starting and inductance method, it can make more accurate positioning and acceleration, find a more suitable back EMF switching point, and have higher starting efficiency and starting success rate. Through experiments and exploration, the feasibility of the algorithm is verified.

In this paper, we present an automatic colorization model for gray scale images based on a generative adversarial networks (GAN). The proposed model is composed of U-Net structure generator and a discriminator with multi-layer convolutional neural network. Then the loss function and weights of true reliability value in the discriminator are optimized. Finally, the corresponding constructed gray scale images colored and colored images are used for training to get an automatic coloring model. In experiments section, we provide various visualization results to validate the effectiveness of our method. The result of a large number of gray scale images coloring experiments show that the model has good learning ability on color transfer and can achieve better coloring effects.

In the actual operation scenario, we usually want to measure the trajectory of the UAV performing the task. This can be achieved by many different means. For example, we can obtain the motion state estimation of UAV by obtaining sensor information such as GPS, RTK and IMU. Alternatively, the velocity and acceleration of the UAV can be measured and its displacement can be calculated by integration. The device (including hardware and algorithm) to complete this motion estimation is called odometry. Considering the efficiency and cost, we hope to complete the devise of odometry visually. The target of visual odometry (VO) is to estimate the motion of the camera according to the captured image. This paper presents a method based on feature extraction and incremental pose estimation, which can accurately calculate motion state of UAVs through aerial images taken by its own.

This paper presents an approach to construct Beijing PSC (Putonghua Shuiping Ceshi) Database by updating and refining the data fields of the original database of Beijing PSC 2020. Study shows that the newly constructed database has met the research needs. By analyzing the data, we find that the overall level of Putonghua in Beijing is high, and it has an overwhelming advantage at the higher levels; Moreover, as the main part of the PSC, college students are the main source of “Level F” candidates, most of whom are from the Northwest and North China. Based on our findings, we suggest that, first, more Putonghua training and testing centers in colleges and universities be established; second, the funding for Putonghua scientific research be increased; third, language intelligence technology in PSC be enhanced.