This PDF file contains the front matter associated with SPIE Proceedings Volume 12248 including the Title Page, Copyright information, and Table of Contents.

Formaldehyde electrochemical sensors are arranged on side walls, ceilings, etc, which can't effectively reflect the concentration level of formaldehyde, so the ventilation system can't respond in time or respond too fast, resulting in the ineffective removal. In this study, we set up an experimental room to monitor the formaldehyde in different locations under different working conditions, then linearly fit the detected concentration to the target concentration. With the same working conditions' average correction coefficients, correcting the actual point monitoring data. The results show that the average error of the monitoring data after the correction is reduced by 46-71%. This method is easy to operate and has certain reference significance for the sensor setting in actual engineering.

Design of a miniaturized synchronous phasor measurement device that can be installed with a large scale and suitable for distribution network lines is of great significance to the intelligent development of the distribution network. This paper designs a miniaturized distribution network line synchronous phasor measurement unit (PMU) based on 5G mobile communication and tests its main characteristics. First, the synchronization phasor measurement algorithm was researched, and a DFT algorithm that uses the hamming window as the window function to improve the measurement accuracy and measurement error of the dynamic synchronous phasor was proposed. The GPS receiver was used as the synchronization signal source to synchronize the clock timing of the PMU. The test results showed that the GPS receiver timing accuracy meets the device clock timing requirements. 5G communication from the PMU to the cloud platform of the distribution master station can meet the communication requirements of the distribution network. The PMU device meets the requirements of miniaturization, high precision, low delay and high timing accuracy for synchronous phasor measurement of distribution network lines, 5G and the Internet of Things can meet the needs of application scenarios in distribution network terminals.

Based on the concept of multi-sensor fusion, this paper proposes an autonomous walking intelligent garbage recycling vehicle with NUC 7i5BNK as the main controller. The recycling vehicle is mainly designed from the system hardware. The main contents of the hardware design include the detection module, control module, and drive module. The detection module is each sensor, which feeds back the real-time status of the recycling vehicle. The core processing chip of the control module is NUC 7i5BNK, which is connected to the embedded connection to obtain the data information fed back by the sensor, fuses the information, and sends instructions to the motor driver board of the execution module to control the operation of the motor. After many actual measurements, the intelligent recycling vehicle can accurately perform autonomous walking and automatic obstacle avoidance. In the process of autonomous walking, the intelligent recycling vehicle has a small output error of attitude angle, high walking path accuracy, stable operation, high flexibility, and wide coverage.

The wireless communication radio frequency transceiver system is the front part of the communication system, and it is also the basis of the normal operation of the communication system. If the transceiver system does not operate smoothly, it will affect the communication effect of people. Therefore, strengthening the design and optimization of the wireless communication radio frequency transceiver system plays an important role. In the actual situation, although my country's communication system has made great progress and various technologies have been continuously improved, there are still many problems in the wireless communication radio frequency transceiver system. When designing a wireless communication radio frequency transceiver system, many factors need to be considered to effectively improve the quality of transceivers. Therefore, the article explores the structure and principle of each module on the basis of the basic principles of wireless communication radio frequency technology. After that, I will fundamentally put forward suggestions for improving the performance of each module to achieve the wide bandwidth and high rate of the wireless transmission radio frequency transceiver system.

This paper presents the design and implementation of a wideband C-band digital array module. The main components of the digital array module, including wideband digital T/R module, frequency synthesizer and power module, are discussed in detail. At reasonable design cost and effort, the proposed module with a relatively compact size and flexible design is fabricated based on field programmable gate array. In this paper, a high intermediate frequency wideband signal is generated based on direct digital synthesizer in a polyphase structure by using coordinated rotation digital computer algorithm. What is more, a wideband waveform real-time delay compensation technique based on variable fractional delay filter is proposed to satisfy the beamforming requirement. At last, simulations and experiments verify the feasibility and effectiveness of the proposed wideband digital array module. The measurement results in this paper will help to a better understanding of the wideband digital array radar.

Due to the different effects of the factors affecting the power communication network security situation on network security, the accuracy of security situation awareness is low. Therefore, a power communication network security situation awareness method based on LDA-RBF is proposed. The linear relationship between influencing factors is established through LDA, and the hierarchical structure model is constructed to normalize the situational awareness factors and divide the corresponding levels of different influencing indicators. RBF is used to construct security situation awareness function, and ReLU is introduced to replace tanh in RBF to improve the accuracy of perception results. The test results show that the design method has a high degree of fit between the network security situational awareness results and the actual value, and has a good application effect.

There are many popular e-book formats on the market, ePub3.0 is very prominent among them, and many software and hardware manufacturers have adjusted their products to be compatible with the ePub format. ePub3.0 has the advantages of being cross-platform, perfectly supporting scripts and supporting the interaction of multimedia contents such as sound and video. Based on the development trend of mobile phone readers and the characteristics of Web App with low development cost, fast iteration cycle, and excellent cross-platform, a web-based ePub e-book reader system is designed. This system will realize the integration of the local service of the Web App mobile terminal and the operation mode of the traditional web application. This system uses Html 5 and Vue 3.0 for client writing, adopts B/S architecture as the entire project architecture, and Nginx provides static resource guarantee for the server. In the early stage of development, according to the current situation at home and abroad, the demand analysis was carried out, and the current development trend of mobile web readers was understood. Finally, a functional test was carried out to test the usability, reliability and practicality of the project.

In the application process of wireless sensor network localization algorithm, the localization accuracy may be affected by structured noise interference. In order to effectively solve this problem, it is necessary to deeply study the existing localization algorithms. In the specific research process, we can make in-depth research on the localization algorithm of wireless sensor networks based on the existing matrix completion localization algorithm and the norm regularization matrix completion localization algorithm, in which the localization of intermediate nodes is an important foundation to realize the application of wireless sensor networks. When analyzing the traditional location method based on measured distance, it needs a lot of node distance information and multipath effect. However, the WSNs node localization method based on norm regularization matrix completion is based on the low-rank characteristics of the distance matrix between nodes in the sensor network, modeling some specific recovery problems such as partial sampling information, and forming the matrix completion problem under the condition of coefficient outlier noise, which can improve the localization accuracy, and at the same time has good anti-interference ability for various sequencing noises.

A stretchable sensor was successfully fabricated by suturing multilayer elastic knitted fabrics using conductive yarn coated with silver in this manuscript. A liner relationship between the resistance of the sensor and the elongation of knitted fabrics in a certain range were introduced. There were no significant wrecks on the surface of silver-coated yarns before and after stitching. Repeated tensile cycle tests showed that the sensor had high repeatability and good stability. The fabricated processes of the sensor were only using the sewing processes of manufacture cloth, so that the sensor could be a partial of our daily garments, which could monitor body movements of sports and other stretch actions of human body, The sensors could be designed as a part of smart garments which could monitor our healthy and movements.

To improve the efficiency of power generation systems, the boiling-like phenomenon in supercritical fluid heat transfer has attracted much attention. However, traditional tapered fiber optic probes are not suitable for supercritical fluid twophase- like distribution detection. In this paper, a fiber optic probe sensor with a flat end face that can continuously measure the refractive index and density of fluids is developed and tested using water, CO₂, and ethanol. The average relative error of density measurement is less than 0.1 %. Applied to a supercritical fluid pool heat transfer experimental system, the probe's signal coincided with the visualization phenomenon very well. The frequency of gas-like film detachment from the surface of the heating wire could be accurately measured using the probe.

In order to solve the essential problem of extending the network lifetime on the premise of ensuring the network quality of coverage, a redundant node decision rule based on binary sensing model and no accurate location information is proposed, and a dormancy scheduling algorithm based on clustering protocol (DSACP) is designed. Furthermore, the dormancy scheduling algorithm can select the set of active nodes and put the redundant nodes into a hibernation state. Simulation results show that the algorithm can select as few working nodes as possible, reduce the energy consumption of sensor nodes and extend the lifetime of the network.

To meet the requirements of temperature controlling of MTBE complex unit, MTBE complex unit temperature control system is designed. The principle of MTBE complex unit temperature control system is introduced. Aiming at the characteristics of the controlled object, the mathematical model of the system is constructed and analyzed. By adopting the method of Feedforward and variable region fuzzy control system, the structure of system hardware and software are designed. Results of experiment indicate that the designed temperature control system has the strong anti-interference performance and robustness. It can meet the requirement of accuracy and stability of MTBE complex unit temperature control system.

This paper designs a tracking system based on Beidou navigation and positioning. The system uses the Beidou positioning module to obtain the position information of the wearer in real time, which can be uploaded to the OneNET cloud platform through NB-IoT wireless communication module. You can also customize the Web on the PC side to view the wearer's current position information and movement trajectory. In this way, the guardian can grasp the location of the elderly in real time, and can find the elderly in time when they are lost, so as to avoid accidents of the elderly. The system is stable in function, simple in operation, convenient in use and accurate in positioning, and can effectively solve the problem of lost elderly people with dementia.

As an important technology of biosensors, Fiber Bragg Grating (FBG) has received more and more attention due to its high sensitivity, fast response, the intrinsic electric safety, the immunity of electromagnetic interference, resistant to corrosion. In this paper, a brief theoretical overview on the sensing principle of FBG biosensors is provided. And various technologies of FBG for improving the sensitivity of biosensors are summarized, such as normal FBG, tilted FBG, long period grating, etched FBG, and FBG biosensor combined with Surface Plasmon Resonance (SPR). In addition, recent advances in biological applications are described, the merits and demerits in detection process are analyzed and compared, and the future development is also prospected.

Internet of things is enabled by a large-scale network which is proposed to provide omnipresent connectivity and a close integration of human and thing, containing dense passive devices and central readers or ambient RF sources. The crucial elements in RFID system are supposed to be tags integrated with proper antennas and digital sensors. In this paper, previous works regarding to RFID concept, sensing materials and backscattering will be discussed, as well as potential application in sensing industry such as food safety, manufacturing monitoring. Based on such review, a proposed solution namely T2T network for resolving conventional RFID system will be presented and numerical results are discussed in the end.

There are some technical problems in the preparation of mixed charge for calcium carbide furnace, such as cumbersome operation, inaccurate ratio and uneven mixing. Based on the analysis of burden proportioning calculation, working principle of batching equipment and batching operation requirements, this paper designs a control system based on frequency converter, PLC and HMI by using frequency converter frequency modulation and master-slave following method, which well meets the requirements of accurate preparation, mixing uniformity and multi batch operation of calcium carbide furnace mixture. The practical application in the production field shows that the designed control system has the advantages of accurate burden ratio, simple operation, high production efficiency and reliability.

When using ultrasonic to measure the dynamic burning rate of propellant during the test of solid rocket motors, it was necessary to grasp the ultrasonic attenuation characteristics of the insulation layer. This paper used the ultrasonic continuous pulse reflection method to obtain the multiple echo data of the insulation layer by setting the appropriate gain multiple, the echo intensity of the ultrasonic wave after passing through the insulation layer of different thickness was calculated and analyzed to obtain the ultrasonic attenuation characteristic curve of the insulation layer. This method can be used to calculate the maximum thickness of the thermal insulation layer tested by ultrasonic energy and improve the accuracy of ultrasonic measurement of the burning rate of solid rocket motors.

The temperature drift correction technology of NaI(Tl) scintillation crystal gamma spectrometer is always the important content of nuclear radiation detection research. The causes of temperature drift are studied in this paper. Many spectral stabilization methods are researched and their advantages and disadvantages are analyzed. The research and analysis of these contents are of great significance to the research of nuclear radiation detection technology and the development and application of detection instruments.

The current group key management methods in wireless sensor networks despise the overhead in the actual use of the protocol while enhancing the security performance. In order to solve the problem of high computational complexity and high communication cost, a group key dynamic management method for wireless sensor networks is proposed in this paper. The hash function is used to select the key, establish the wireless sensor network group structure, and generate the leading node by election for identity authentication and data processing. A key agreement protocol is proposed in the group to establish a secure communication channel for terminal groups distributed among different domains. The node obtains the public key of other nodes according to the identifier, mapping algorithm and matrix of the communication partner, and updates the key only for the group users who join or leave, so as to reduce the traffic and computation. Set the threshold value for the number of member nodes. When the number of sub key components obtained by a node reaches or exceeds the threshold value, the cluster key can be recovered and saved, and then other sub key components can be deleted. The simulation results show that compared with the key management method based on symmetric polynomial and logical key tree, this method uses less communication overhead and storage overhead, reduces the time overhead and improves the communication efficiency of the management protocol.

Aiming at the ill-posed problem of the solution in the image reconstruction process caused by the soft field effect of Electrical Capacitance Tomography (ECT) technology in the process of aviation kerosene detection, a hybrid minimization constraint model based on total variation joint sparse regularization is proposed. Applied to ECT image reconstruction, it can not only alleviate the artifact gradient in the smooth area of the reconstructed image, but also accurately approximate the optimal solution with appropriate sparse coefficients, and maintain the sharp edge of the reconstructed image. Since the multi-regularization brings more parameters, the Alternating Direction Method of Multipliers (ADMM) is selected to iteratively seek the optimal solution of the ECT inverse problem. The experimental results show that: compared with traditional algorithms such as Landweber and Tikhonov, the image reconstruction error using the algorithm in this paper has been significantly reduced, and the correlation coefficient has also been improved (in terms of the reconstruction effect of the core stream, the image reconstruction error is reduced to 0.03, The correlation coefficient is increased to 0.96), and the reconstructed image quality of different flow patterns is improved to a certain extent. The experimental results show that the ADMM algorithm based on the total variation joint sparse regularization mixed minimization model can effectively and accurately detect solid impurities in aviation kerosene.

In view of the low measurement accuracy of traditional mechanical energy meters, the inability to charge by time, and the inability to display relevant parameters, a smart energy meter was designed. The smart energy meter takes the single chip microcomputer as the main control chip, adopts the ADE7755 module to realize the collection and calculation of the electric energy, and combines the 12864 display module, the independent button module, the clock DS1302 module, the storage 24C02 module, etc. to improve the electric energy measurement accuracy, realize time-sharing billing and Parameter display function. Completed the hardware design, software design and physical development of the smart energy meter. The physical test results show that the smart energy meter can realize dual-rate charging, reduce resource waste, and the rate range can be adjusted according to the actual situation, and can display data such as electric energy, electricity cost, time, and date.

A metasurface polarization conversion device is proposed by etching a specially shaped hole in the center of a single disk of graphene. The polarization converter can achieve the cross-polarization conversion function from x to y in the midinfrared band, with three polarization conversion rate peaks and a maximum conversion efficiency of more than 98%. By adjusting the chemical potential of graphene, the operating frequency can be dynamically adjusted to ensure that the conversion efficiency of two operating bands is above 90% and that of three operating bands is above 80%. Based on the above performance, the designed polarization conversion device can be potentially applied in the field of optical polarization control.

As a weak magnetic sensor, optical pump magnetometer is widely used in resource exploration, space exploration, and other fields. According to the principle of measuring weak magnetic anomaly signal under geomagnetic field, a rubidium optical pump magnetometer experimental system is built based on the methods of laser pumping and RF(radio frequency) field depolarization, The influence of RF field intensity on the resonance spectrum of rubidium optical pump magnetometer is studied. The results show that the intensity and linewidth of the resonance spectrum increase with the increase of RF field intensity, but the ratio of linewidth to intensity has a minimum. This paper provides a technical basis for the sensitivity optimization of rubidium optically pumped magnetometer.

Aiming at the problem of high runoff coefficient caused by unreasonable pipeline layout in the whole drainage system of sponge airport, a whole drainage system of sponge airport is designed based on BIM. In the hardware, the data acquisition and power supply module is designed by using the main control chip, a variety of sensors and their front-end circuits to complete the voltage conversion. In the software, according to the whole structure of the sponge airport drainage system, this paper determines the rainwater treatment facilities and their combination, and calculates the scale of the drainage facilities. In this paper, the location of BIM is selected, the road layout and pipeline distribution are integrated, and the whole drainage model is established. The design rain pattern of the system test adopts Chicago rain pattern, the rainfall duration is 2h, the rainfall peak rainfall coefficient is 0.4, and the return period is 10 years. The results show that the average runoff coefficient of the system designed in this paper is 0.403, which is 0.374 and 0.376 lower than the whole airport drainage system based on LID and SWMM. Therefore, the system can significantly reduce the airport runoff and has high practical value.

Although the performance changes of 25Cr2NiMo1V steam turbine steel have been studied through experiments under high temperature creep conditions, the performance degradation caused by long-term service is difficult to clarify in the laboratory due to the more complex actual service conditions than the experimental conditions. The influence of service conditions on the mechanical properties of 25Cr2NiMo1V rotor steel has not been clearly explained. In order to investigate the performance degradation behavior of 25Cr2NiMo1V rotor steel under long service, the mechanical test of in situ compression was carried out on the micro-nano scale in the base metal area of the welded joint of 25Cr2NiMo1V turbine Sub-steel which has been in service for more than 20 years. The results show that the young's modulus of the sample after long term service is about 77.1% lower than that of the original material; The strengthening index of the sample is about 3.6-6.3 times higher than that of the bulk sample. Due to the influence of local microstructure, the yield strength data fluctuate greatly. Considering the size effect, it can be considered that long-term service has an obvious adverse effect on the yield strength. The current study provides some critical insights into the reliability evaluation of long-term service steam turbine rotors from the micro-scale.

The realization of high-precision of the concentration of PM_2.5 is of great significance for guiding people's traffic travel, taking targeted intervention measures, and establishing an early warning and forecasting system. However, due to various factors, it is difficult to achieve high-precision prediction. In order to improve the prediction accuracy of PM_2.5, this paper proposes a deep learning prediction model based on Lasso for dual feature extraction and Multilayer Perceptron (MLP). Firstly, on the basis of data analysis and processing, the number of leading steps is selected by LASSO, and the hyperparameter is optimized based on Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) (hereinafter referred to as "LASSO-LarsIC"); Secondly, on the basis of the selection of leading steps, LASSO-LarsIC is used to extract features of different influencing factors in different leading steps, 76.92% of different influencing factors at the same leading step are eliminated, which reduces the dimension of input variables and improves the generalization ability greatly; Finally, the extracted optimal feature set is used as the input variable, and MLP is applied to predict the concentration of PM_2.5 with 1 hour in advance. The verification results show that the prediction accuracy deviation in this paper is 4.85μg/m³, and improved by more than 30% before and after feature extraction, the effect of feature selection is obvious.

In order to explore the effect of low temperature plasma reactors on the degradation of organic wastewater, a plate dielectric barrier discharge reactor was designed with copper foil as electrodes and quartz as a dielectric barrier. Under the output voltage of 13kV, 4mg.L^-1 methylene blue solution was used to simulate organic waste liquid and degradation. The width X of the discharge electrode, the gap D between two discharge plates, and the height of the discharge distance from the liquid surface H were changed to explore their effects on the degradation of methylene blue. It was found that the smaller electrode width its electric field density is high and easy to discharge that had better degradation of methylene blue. A smaller discharge space is not easy to transfer the discharge products, will cause accumulation, saturation, and reduce the decolorization rate of the methylene blue solution. The height of the discharge electrode from the treatment liquid surface will affect the life of the active substances, resulting in a decrease in the decolorization rate of methylene blue. All research results provide necessary reference factors and new ideas for the design of plasma reactors to treat organic wastewater.

The operation environment of transmission line directly affects the operation safety of transmission line. In this paper, a power tower inclination sensor based on Fibre Bragg grating is studied. Firstly, the packaging method is studied. A heavy object is hung at the free end of the equal strength cantilever beam, and an FBG is pasted on the symmetrical centreline on both sides of the equal strength cantilever beam. When the power tower tilts, the weight drives the free end of the cantilever beam with equal strength to produce deflection, causing the FBG to shift the central wavelength. The tilt angle measurement of the tower can be realized by monitoring the central wavelength. Then the corresponding installation fixture is designed. Finally, the performance of Fibre Bragg grating tower tilt angle sensor is tested, and the test results show that its performance is good.

Colorectal cancer is one of the digestive tract malignant tumors in life. In recent years, the incidence rate of rectal cancer has increased significantly in China. The segmentation of CT images of colorectal tumors is a tedious, time-consuming and expensive job.When diagnosing rectal cancer, it would be helpful to accurately segment the rectal tumor region from CT images so that doctors could make more accurate diagnoses and thus have better results for the disease. Based on the literature review, this paper presents medical image segmentation algorithms, including traditional algorithms based on energy minimization, and evaluates them according to the metrics used. The paper also describes the public datasets available for use in the study. In this paper, various segmentation methods are analyzed in detail. Their theoretical and practical characteristics are described for each group of methods, and the most advanced and promising methods are proposed. This paper can help understand the available colorectal segmentation methods, making developing new and more effective and improving existing methods easier. The method is simple to interact with, robust and accurate, and can effectively assist physicians in achieving diagnosis and surgical planning.

In brand design, enterprises should make full use of the era background of big data of the Internet of things and make a comprehensive and systematic analysis of space, color and sound from the perspective of visual communication. Firstly, this paper introduces the big vision software technology and the research situation and development trend at home and abroad. Secondly, it discusses how to integrate the three-dimensional model and digital interactive media expression form by using the professional method of art design based on big data. Through the use of questionnaire survey method, this paper investigates the brand design tendency choice of consumers of different ages. Finally, the survey shows that young people aged 15 to 30 tend to be young and fashionable, while those over 35 are more practical and simple.

With the rapid development of society and the continuous innovation of the times, the Internet has gained a lot of room for development. Our country has initially entered the information age. When conveying information, it is mainly through visual transmission, which provides more possibilities for the fast-developing society. This article will combine the development characteristics of network media to study the design of visual communication in the network media environment. This is a brand-new way of communication that improves people's quality of life to a certain extent. The design concept also needs to be innovated, based on the network media environment, to develop a new visual communication design. This paper will conduct the in-depth study of the visual communication design in the network media environment from three aspects.

In order to solve the problem of the low efficiency of the traditional slicing method in measuring the volume of objects with continuously changing surfaces, a volume calculation method based on improved point cloud slicing is proposed in this paper. This method uses the product of the absolute value of the difference between the areas of adjacent slices and the corresponding spacing of adjacent slices to represent the magnitude of the object's surface variation, adjusts the spacing of slices according to the magnitude of the variation, calculates the area of each slice using the proposed Monte Carlo method, and finally uses the idea of integration to find the volume of the target object. The experimental results show that the improved point cloud slicing method can improve the calculation accuracy of the traditional slicing method without increasing the number of slices, and can better solve the volume calculation problem of objects with continuous surface changes, which has good application value.

With the continuous advancement of Internet construction, the scale of enterprise communication synchronous optical transmission network continues to expand and the network structure is becoming increasingly complex. There have been problems such as reliability, reduced survivability, and difficulty in adapting to development needs for operation and maintenance management that only relies on personal experience and theoretical analysis. This paper designs a routing weight function for the joint balance of risk and load, and proposes a network service routing planning algorithm that integrates multiple constraints based on reinforcement learning Q-Learning. Experiments verify the effectiveness of the proposed algorithm, which can reduce the congestion rate and network business risks in the context of balancing network risks and load flows. It can realize the business routing planning and simulation prototype system of enterprise communication.

Text classification is the most fundamental and essential task in natural language processing. The last decade has seen a surge of research in this area due to the unprecedented success of deep learning. Previous models assume that all classes are equally difficult to distinguish and treat all of them exclusively. But in fact, the property of general-to-specific category ordering often exists between classes. In this paper, we exploit the prior knowledge of class hierarchical structure to enforce the network to learn human-understandable concepts in different blocks and propose a new model named H-CRNN, which combines TextCNN and Bi-LSTM to construct a hierarchical structure. We test our proposed model on the THUCNews dataset, and experiments show that our proposed H-CRNN model achieves the best results than other methods.

With the rapid development of social economy, users have higher requirements for power quality. As the last level of power system, the reliability research of distribution network has been paid more and more attention. However, due to the random failure of system components, some or even all of the system functions are lost, which brings more and more huge economic and social losses to the normal production and life of modern society. This paper analyzes the topology of distribution network by using deep neural network, reveals the influence of network topology similarity on reliability from the perspective of network graph theory, and obtains regular conclusions. On this basis, the failure rate level of key branches, grid structure and access location of distributed power generation are further evaluated. The mathematical relationship between network topology and reliability is described with the index as the medium. Finally, the linkage optimization of reliability index is realized through the optimization of the above indexes, which solves the unclear optimization objectives in traditional planning. It pursues local optimization and ignores global optimization. Experiments show that the research content of this paper can effectively evaluate the health of power grid structure, and provide a reference for the research of distribution network graph theory.

As a main direction of the new generation of mobile communication technology, 5G can not only bring the user experience of ultra-high bandwidth, ultra-low latency and ultra-large-scale connection, but also provide basic support for future economic and social development. Nowadays, many storage yards are undergoing intelligent transformation by adopting remote control system. However, the operation process of the storage yard will stagnate when an emergency occurs. And the traditional communication methods have low efficiency and high maintenance costs. Aiming at improving the efficiency of the storage yard, this paper adopts 5G network on the remote control of yard crane, and a mixed-integer linear programming model is established. The branch and bound algorithm is adopted to solve the model on Lingo17.0. The experimental results show that the comprehensive application of 5G network and yard crane remote control technology can improve the efficiency of the storage yard.

In this paper, the low noise design scheme of MEMS gyroscope for north-finding is studied. Firstly, the system level noise model of MEMS gyroscope is established, and the low noise design is carried out around the structure and circuit. In terms of structure, the design method of low mechanical noise and large capacitance sensitivity is studied. In terms of circuit, firstly, the noise generation mechanism and suppression scheme of each module are studied. Secondly, the charge injection scheme is used to compensate the quadrature error, and the zero-bias instability of the gyroscope prototype is obtained as 0.12°.h . The four-position north-finding scheme was used for the experiment, and the statistical standard deviation of the north angle was 0.298° within 4.5 min.

In hydraulic engineering, as a low head hydraulic structure, sluice plays an important role in regulating water resources, agricultural irrigation, safe flood control, industrial production and so on. Due to the limitation of geological conditions, many sluice gates are built on soft soil foundation, which usually can not meet the requirements of foundation bearing capacity in the structural design code. In practical engineering, pile groups are usually designed under the sluice bottom plate to make the pile foundation and foundation jointly bear the upper load. In this paper, taking the actual project of sluice as an example, based on the ANSYS finite element simulation technology, considering the dynamic interaction of water body structure foundation and the nonlinear problem of pile-soil contact, the finite element simulation model is established and analyzed. The dynamic coupling system of structure and water body is established. The natural vibration characteristics of the overall structure are analyzed. And the numerical analysis results are compared with the analysis results of measured data. The calculation results of the finite element model established according to FSI theory are more consistent with the modal identification results of CEEMDAN-SVD. The calculation accuracy and modal identification are better. They confirm each other and can more accurately reflect the working characteristics of the sluice structure. In the working state, the modal identification method can accurately extract the modal characteristic peak of sluice structure, and can be better applied in the dynamic characteristic analysis of sluice structure.

5G technology has opened a new era of the Internet of everything, and 5G+ application has become the core focus. Under the background of accelerating 5G construction, in order to meet the core needs of smart library construction, this paper uses 5G technology to empower the smart library, constructs a four in one "5G + smart library" comprehensive management system of "perception + network + data + application", and designs a 5G network, and puts forward the application scenarios of 5G-driven smart library. The comprehensive management system applies 5G technology to the construction of smart library. It can effectively solve the bottleneck problems faced by the library, turn the traditional library to "smart library", enrich the new experience of digital reading, and better meet the diversified and personalized reading needs of users.

Improving retrieval accuracy and matching efficiency are the two main goals of shape retrieval. In order to improve the accuracy of shape retrieval and the efficiency of matching, it is necessary to design a compact and discriminative shape descriptor. The Multi-Scale Arch Height (MSAH) shape descriptor has good retrieval performance, but the description of the shape is still insufficient. Therefore, a multi-scale Fourier shape descriptor based on arch features is proposed. The proposed shape descriptor fully describes the global characteristics and local details of the shape. While maintaining high retrieval accuracy, it also has high computational efficiency. We tested the proposed descriptor on MPEG-7 image library. The results demonstrate that the performance of the proposed descriptor is better than the Multi-Scale Arch Height (MSAH) shape descriptor and other classic Fourier descriptors.

The location of three-dimensional coordinates of the target is a very important research in the fields of radar, electronic countermeasures and wireless communication. It plays an extremely important role in the drawing of topographic maps, environmental monitoring, disaster prevention and modern warfare. At present, there are problems that the location method is complex and the location accuracy is not high. Aiming at this, this paper proposes a method to locate the strong scattering point by using SAR image, which is combined with the range Doppler model to locate the target. Analytical solutions are derived from the theory of equations and the error factors are analyzed and discussed. Finally, an experiment was carried out to locate the five corner reflector targets with strong scattering characteristics using synthetic aperture radar, and the three-dimensional coordinates of the strong scattering point targets were obtained. The experimental results demonstrate the reliability of the method.

Advances in computer technology and artificial intelligence have led to a surge in the volume and complexity of multimedia data, and content-based image retrieval systems (CBIR) have become widely popular to extract useful information from this data. Traditional CBIR systems are implemented by inter-image features and cannot perform associative image retrieval. Therefore, in this paper, we propose a new method for associative image retrieval by mimicking the human brain mechanism. By using an improved deep residual network to extract different kinds of image features, and using a loss function to maximize the feature distance loss between non-associative image groups and minimize the loss between associative image groups, and finally using the trained model to achieve multi-objective associative retrieval.

To solve the problem of low accuracy of detection and positioning of invading foreign body targets of the scene of railway foreign body intrusion, this paper proposes a railway foreign body intrusion detection method that optimizes the YOLOv4 model. First, the self-built railway foreign body intrusion limit dataset was expanded to solve the model over-fitting problem and improve the generalization ability of the model. Second, replace the backbone feature extraction network in the YOLOv4 model with a lightweight feature extraction structure, which effectively reduces the total number of parameters and deepens the number of network layers. Experiments show that the algorithm has good results and performance in detection accuracy and detection speed, effectively solving the problems of missed detection, misdetection, and low detection accuracy of foreign objects in the current road foreign object intrusion scene.

Pointer meters are widely used in transformer substations and other places because of their good robustness. Generally, the reading of pointer instrument cannot be read automatically, and can only rely on manual reading. With the development of robot and computer vision technology, it is possible to use inspection robot to obtain the reading of pointer instrument. The automatic reading of the analog meter should first detect the meter from the image, and then identify its reading. For the instability of traditional instrument detection, an analog meter automatic reading method based on YOLOv5s detector is proposed in this paper. Hough transform algorithm is used to detect the pointer and range, and then the meter reading is realized based on the angle method. Experiments show that the average processing time of a single image is 0.3s, and the reading error is less than 5%, which can meet the requirements of automatic reading in patrol inspection.

Aiming at the problems of long operation cycle and low system operation efficiency in the two-level priority exhaustive polling service system, this paper proposed a two-level priority polling system based on parallel exhaustive services mode. First of all, this article combines the idea of parallel optimization with the polling system, changing the traditional pipeline scheduling method, and optimizing the transfer process and service process in the polling system in parallel, which improves the efficiency of polling. Then, through the mathematical method of probability generating function and Markov chain, the two-level exhaustive polling service system is analyzed mathematically, and the first-order partial derivative of the probability generating function of the system state variables is obtained to obtain the equations of its characteristic parameters, which are solved simultaneously get the complete mathematical expression of the average queue length. Finally, the computer simulation experiment analysis of the system is carried out, and the experimental results are consistent with the theory, which proves the reliability of the theoretical analysis.

The accuracy of spectral recovery is directly affected by the spectral filter. Clarifying which parameters of the color filter will affect the spectral reconstruction accuracy can help us select the appropriate color filter and improve the reconstruction accuracy. Gaussian curve is used to simulate the spectral sensitivity curve of camera CCD channel. Taking color difference, goodness of fit coefficient and root mean square error as the result evaluation parameters, three, five and seven spectral filters with bandwidth of 20-60nm are selected for experiments. The influence of the number and bandwidth of camera channels on the spectral recovery results is explored by calculating the spectral recovery accuracy of Munsell spectral data set. The experimental results show that the number and bandwidth of spectral filters will affect the spectral recovery results.

For the problem of traditional 16-beam LiDAR has low detection accuracy for obstacles on the road, this paper developed an improved DBSCAN clustering algorithm, which can dynamically select the threshold parameters according to the distance between the point cloud and the LiDAR, so as to detect obstacles quickly and accurately. Fistly, according to the geometric features of the point cloud, the road curb points are extracted and the left and right roadside lines are fitted. Then, the region of interest is extracted based on the two lines. After the ground segmentation on the region of interest, the improved DBSCAN clustering algorithm is applied to complete the detection. Experiments show that this method can cluster the near and far obstacles quickly and accurately.

With the development of economy, electric bicycle has become a common tool for people to travel. However, the theft of electric bicycles is common, especially batteries, which are easy to be stolen. To solve the above problems, an intelligent anti-theft system for electric bicycle is designed. The system detects the vehicle state through a variety of sensors such as vibration and positioning. Under the function of the system, once the electric bicycle has vibration, displacement, battery loss and so on, the owner will be notified through the network for prevention.

There are a lot of works needs to be done for determining the safety status of escalator or moving walk. For example, at least 100 risk assessment projects are need in determining the whole machine safety status. Besides, the data is used repeatedly in generation of safety assessment report. In order to solve the above problems, a safety assessment of escalator or moving walk software is provided in this paper. The software is programmed based on Visual Basic, Word and Excel. Risk assessment of every single project can be got in a selective way. The safety status of the whole Escalator or moving walk can be computed automatically based on the risk assessment results of every single project. The safety assessment report can be generated automatically, and the data generated in the safety assessment process can be used repeatedly without much effort.

At present, the target detection algorithms represented by deep learning, such as: You only look once (YOLO), Single Shot MultiBox Detector (SSD) and other single-stage target detection algorithms have demonstrated high detection rate and stable detection effect However, when this type of detection algorithm detects certain types of targets in remote sensing images, because the relative and absolute sizes of these targets are very small, the problems of low detection rate and high false detection rate will occur. Traditional detection methods cannot meet the requirements of remote sensing. Image inspection needs. To solve this problem, this paper proposes a small target detection method based on image enhancement method. Firstly, the image overlap and block method are used to increase the relative size of the small target and retain the information of the small target. Secondly, the image after each block is The block uses an image enhancement method based on SRGAN to improve the resolution of the small target, which is used to improve the sub-table rate of the small target and enhance its feature and context information. Then based on the YOLOv4 method, the enhanced image block is detected. Finally, based on the NWPU VHR-10 data set, the effectiveness of the method is verified in the experimental results of remote sensing image based on image enhancement of small target detection.

In contemporary society, with the increasing difficulty of tasks required by robots, multi-robot system has gradually become a research hotspot. As multi-robot task allocation is an important part of multi-robot system cooperation, it has also become a hot topic of research. In this paper, particle swarm optimization is used as the basic optimization tool to study the optimization solution of multi-robot task allocation problem. The multi-robot task allocation problem solved by the improved particle swarm optimization, the three main control parameters of the particle swarm optimization are updated by the linear parameter updating rule. Finally, the MATLAB simulation verification is given. The simulation results show that the algorithm has good adaptability in solving the multi-robot task allocation problem.

The detection of road vehicle targets is an important research direction in the field of target detection. As there is some problem in vehicle target detection while using YOLOv3, such as high miss rate and low detection accuracy, an improved multi-scale feature fusion target detection algorithm IFFC-YOLOv3 is proposed. Firstly, optimizing the multi-scale feature fusion based on the YOLOv3 algorithm to improve the detection of small targets while ensuring the detection capability of the network for medium and large targets. Secondly, using K-Means++ target frame clustering on road target detection turns out to obtain new candidate frames to improve its accuracy. Finally, the CIoU optimized loss function is introduced to further improve the accuracy of vehicle target detection. In this paper, the effect of the improved IFFC-YOLOv3 algorithm is verified by means of a personal datasets. The experimental results show that the IFFC-YOLOv3 algorithm achieves good detection results in terms of reducing the rate of missed detection and improving the detection accuracy.

It is of great research significance for the deformation of the airtight ring to measure the air-tight of the aircraft and determine the attitude of aircraft in launching device, as the airtight ring is a part of the launch device. Based on the full analysis of various displacement testing techniques and the actual working environment of the airtight ring, a displacement sensor which can be applied to test the dynamic deformation of the seal ring is designed. On this basis, the test of the static and dynamic characteristics of the sensor were carried out. The result shows that the precision of the designed displacement sensor could meet the test requirements, and can be effectively applied to the dynamic displacement measurement of the seal ring.