This PDF file contains the front matter associated with SPIE Proceedings Volume 12127, including the Title Page, Copyright information, and Table of Contents.

Many precision components, such as silicon wafers, optical lenses, etc., require high surface quality for their working performance. In the process of picking up and transferring components and materials, the surface of the components is easily damaged by physical contact. The high requirements of modern precision instruments and equipment on the surface accuracy of components highlight the limitations of traditional contact picking and transfer methods. Based on the nearfield levitation technology and the principle of vacuum adsorption, a new type of non-contact pick-up and transfer device is proposed in this paper. It can be used as an end effector of robotic arm for the pick-up and transport of disc-shaped precision components. The device realizes non-contact operation of disc objects with the action of the grasping suction force provided by the negative static pressure and the levitation repulsion force generated by the ultrasonic field. Based on theoretical and simulation methods, the structure of the non-contact grabbing and transporting device is designed, and the working process of the levitation grabbing of the device is analyzed in this paper. A non-contact grasping operation experiment testing device is built, and the grasping experiments to disk-shaped objects are carried out. Experimental result demonstrates that the proposed device can realize the non-contact effectively and be reliable picking of disc-shaped objects.

A great deal of dust is produced in the process of coal drawing at the working face. Efficient atomizing and dust removal device is the key to effective dust reduction at the working face. Because of the defects of poor atomization effect and easy blockage of the traditional atomizing nozzle in the fully mechanized top coal caving face, propose a highly efficient atomizing nozzle structure, based on computational fluid dynamics (CFD), the outlet diameter and inner diameter of swirl core of a new swirl core nozzle were simulated. The simulation results show that under the same inlet pressure, when the nozzle outlet diameter is 1.5 mm, the outlet velocity is the largest and the atomization effect is the best; Properly reducing the inner diameter of swirl core can enhance the swirl intensity, increase the dust collection area and improve the dust reduction effect.

In order to ensure the safe, efficient and low-noise operation of the Roots blower, this paper is based on a preliminary design of an air filter and muffler integrated device including an air filter and an air intake muffler. The FEM method and CFD method are used to numerically simulate its acoustic performance and aerodynamic performance. Taking the transmission loss and pressure loss at the inlet airflow velocity of 21m/s as the optimization goals, under the premise of without changing its external dimensions, the main internal structural parameters such as filter element thickness, insertion method of intubation, diameter of perforated pipe, and effective perforation length are optimized by orthogonal test design method. The sequence of structural factors that affect the overall performance of the air filter and muffler integrated device is obtained, and the optimal combination of main structural parameters is determined. Based on the optimal model obtained by the orthogonal test, the design of the length of the intubation of the air filter and muffler integrated device is optimized to obtain the best combination plan. The results show that the structural parameters are optimized through orthogonal tests, the overall performance of the air filter and muffler integrated device after optimization is significantly improved, the amount of overall noise reduction is increased by 18% , and the pressure loss is reduced by 20% on average compared with the original model.

Aiming at the current courier companies' security inspection coding generally adopts manual coding operation, the coding speed is low, and the coding quality is guaranteed, a fast coding device based on color characteristics and rapid segmentation by K-method is designed. System, and image processing are composed of two image systems. The image processing system is used to obtain the coordinates of the express location, and the machine executes the operation of fast and high-speed coding. The express positioning algorithm first performs image color feature processing on the collected express, performs Gaussian filtering of noise, uses K-means to shoot the target, performs critical value segmentation of the target, morphological processing and minimum contour detection to extract the diagnosis center coordinates, and finally the extracted coordinates are passed to the mechanical result agency for coding. Guarantee: Using the inkjet code designed in this article to carry out the courier security inspection coding experiment, the accuracy rate reaches 98%, which greatly improves the express security inspection device.

For improving the stability of the liquid supply, the pulsation of pressure and flow, and the hydraulic impact in the operation of the emulsion pump station, a simulation model is established to analyze the impact of the relevant parameters of the accumulator on the stability of the system, such as the various configuration schemes of the accumulator station, and inflation pressure changing of the accumulator. An optimization method based on accumulator theory research and engineering practice is proposed to reduce noise and extending the service life of multiple components and promote safe and efficient production in coal mines. The results show that each pump is equipped with one 20 L accumulator in which the inflation pressure is about 70% of the system’s working pressure. The emulsion pump station equipped with an accumulator station which is composed of eight accumulators in parallel, the volume of each accumulator is 63 L and the inflation pressure is 70%～80% of the system’s working pressure. At this time, the stability of the emulsion pump station system is better.

Soft robots have good adaptability to the environment, good structural softness, high flexibility and other characteristics, which have broad application prospects in medical care, irregular or fragile object clamping, rescue, archaeological investigation and other fields. The terminal execution device of software worm simulation is studied, according to the motion principle of worm, a soft worm-like actuator with bending deformation is designed, the actuator can be combined with a rigid structure and can be applied to related special occasions. In the soft part, a fan-shaped three cavity structure is designed to study the influence on the bending performance of the soft actuator from three aspects: the width of the single-layer air bag, the air bag interval and the air pressure value of each cavity. ANSYS Workbench and MATLAB are used for simulation analysis, so as to design the optimal performance of the software execution part, which provides a theoretical basis for the physical production.

In order to further analyze the classification mechanism of double vortex finder hydrocyclone, the effects of feeding speed and mass concentration on the flow field distribution and classification performance of the double vortex finder hydrocyclone were studied by numerical simulation. The results showed that the increased of feeding speed has no obvious effect on the internal and external overflow split ratio, but it will reduce the bottom flow. With the increase of mass concentration, the overflow fine particle content decreased. Through the classification simulation with different feeding speed and mass concentration, The optimal operating parameters of the double vortex finder hydrocyclone are 9% mass concentration and 11m/s feeding speed.

With the construction of smart mines, intelligent detection methods and technologies have become particularly important. An intelligent detection and early warning system has been proposed for leaks caused by factors such as corrosion, aging, and unintentional destruction of underground coal mine pipe networks. The negative pressure wave detection method was used to solve the difficult detection of complex environmental background, and MATLAB was used to extract noise features to improve operational efficiency. The generalized cross-correlation time delay estimation was not used to improve positioning accuracy. use the theoretical basis of deep sparse filter learning to identify leaks, and achieve accurate classification and identification of leaks by quickly and effectively extracting data features; use fixed base stations under mines as points to implement rapid LAN publish, So that relevant persons in charge can grasp the leaked information and take different response measures in a timely manner.

Rolling bearings are the most common and easily damaged link in mine fans, and there are many problems that can be improved in the acquisition and analysis of vibration signals. Through the establishment of a mining-based rotating machinery failure test platform, the bearing is subjected to fault simulation experiments and the collected signals are processed, and the collected bearing signals are compared without noise reduction processing and the use of noise reduction fusion spectrum algorithm and wavelet noise reduction method After noise reduction, the analysis results are processed by Hilbert transform and EMD (empirical modal analysis), and the frequency domain diagram obtained by comparing the frequency domain map is compared to observe the frequency of abnormal vibration to further infer the fault type. By comparing the analysis results obtained by different methods, they are compared and summarized, so as to promote the upgrading and improvement of processing methods, make accurate judgments on fan bearing faults and give feasibility opinions.

Aiming at the current wear detection of vertical shaft rigid tanks, which cannot directly know the wear value and directly observe the operating conditions of the tanks, a rapid detection method for the wear detection of the tanks is designed. Use Halcon image processing technology to complete the image processing of the tank road image information collected by the industrial camera, and integrate image preprocessing, shape-based template matching and geometric measurement algorithms into the Qt Creator platform, using the UI interface on Qt Creator Realize the functions of display, data storage and early warning of tank wear data. The reliability of the detection technology is verified through experiments, and the experimental tests show that the detection technology can be used for actual detection.

After the grinding process of the piston pin, the outer diameter of the piston pin will be detected by full inspection. The detection requirements shall not produce scratches, the detection rate shall reach 1s / piece, and the limit error of the detection equipment shall reach 0.7μm. Therefore, non-contact high-precision detection must be used to meet the requirements. In order to improve the detection accuracy, carbon fiber tube is used as the reference plane to protect the surface of piston pin from scratches and improve the rigidity of measurements. The measurement error is analyzed, and the overall error is divided into three parts: the error caused by temperature change 𝑒_t , the error caused by the non-perpendicularity of the light and the piston pin 𝑒_v and the error caused by CCD itself 𝑒_d. After adjusting the parallelism between the carbon fiber tube and the piston pin, 𝑒_v is reduced to 0.12μm，a calibration measurement method is developed to reduce 𝑒_d to 0.15μm. A temperature compensation algorithm is developed to reduce 𝑒_t to 0.2μm. Thus the overall detection accuracy is limited to 0.5μm. Finally the high-precision on-line detection of piston pin outer diameter is realized, and the effect is good.

Aiming at the mitigating effect of the damping groove structure of the port plate on the oil backflow and pressure overshoot in the working process of the axial piston pump, a dynamic model of the double compound axial piston pump is established based on the flow area model of the triangular damping groove, and the influence of the structural parameters of the inlet and outlet triangular groove of the port plate on the pressure and flow characteristics of the piston chamber is analyzed. The results show that, as the flow area between the inlet triangular groove and the piston chamber increases, the positive overshoot pressure and the dead volume residual pressure of the piston chamber in the closed pre-release pressure zone decrease, and the oil backflow from the piston chamber to the low pressure area decreases. The phase angle of oil backflow from the high pressure area to the piston chamber decreases with the increase of the length of the outlet triangular groove, and the peak value of oil backflow decreases. The peak value of oil backflow increases with the increase of the width and depth of the outlet triangular groove, while the phase angle of the peak decreases. Therefore, a reasonable configuration of the structural parameters of the triangular groove of the port plate can effectively alleviate the pressure overshoot and oil backflow, and this paper provides a basis for the optimization of the triangular groove.

The mixing oil loss of planetary gear train is affected by many factors, and the flow field of splash lubrication is complex, which is difficult to study directly by theoretical or experimental methods. Therefore, based on MPS method, the planetary gear train transmission was taken as the research object to analyze the power loss of planetary gear train under different speed, oil immersion depth and oil temperature conditions. The results show that the power loss increases with the increase of gear speed, oil immersion depth and lubricating oil temperature, but the effect of lubricating oil temperature on the oil loss is relatively small. It provides a simple, efficient, meshless and accurate simulation method for the analysis of mixing oil loss of planetary gear train.

In order to improve the quality and efficiency of processing and reduce of cost of the micro holes. The two-degree-of freedom axial-torsional coupling structure dynamic model is established, considering the regenerative chatter phenomenon during the machining of large aspect ratio holes based on the macro drilling dynamics model. The nonlinear time-delay differential dynamic equation is obtained, which corresponds to the above model. The dimensionless parameters and variables are introduced to making linearity and dimensionless of the kinetic equations. The different dynamics parameters are compared and the corresponding stability limit diagrams are obtained to predict stable areas. The change trend of the stability area under various parameter is analyzed. The prediction model was verified theoretically through the time-domain response, and the influence of dynamics parameters on the machining stability was also further analyzed.

In order to solve the problem of self weight, the lightweight design of the frame of the tractor manure spreader is carried out. Firstly, the SolidWorks software is used to create a model of the frame, and the three-dimensional model of the frame is loaded in ANSYS software. On this basis, the stiffness and strength of the frame of the tractor manure spreader under full load condition are analyzed through statics, and the weak and redundant parts of the structure are obtained, Thus, the frame structure of the tractor manure spreader is optimized. On the principle of meeting the original function, the total mass of the optimized frame is reduced by 12.743%. Then the modal analysis of the optimized frame is carried out, and the optimal design result is ideal. Structural lightweight design is of great significance to the structural improvement of the frame of the tractor manure spreader.

Aiming at the problem of low accuracy in fault diagnosis of mechanical bearing in coal mine. A method based on CS-fuzzy neural network and wavelet packet transform is proposed for bearing fault diagnosis. Firstly, the collected data is processed by wavelet packet decomposition and reconstruction, and the fault feature vector is established. Then, the CS-fuzzy neural network based on improved cuckoo search algorithm is used to identify and diagnose the bearing fault type. Through experimental verification, the combination method of wavelet packet decomposition and cs-fuzzy neural network can diagnose bearing fault more accurately and quickly.

When the user wears the lower limbs to help the exoskeleton walk, there is bound to be human-computer interaction. If the following performance of the lower limb booster exoskeleton is not good, there is a large error, lag or delay, which will affect the user's action, and even fail to achieve the boosting effect. In this paper, the unilateral lower limb of the exoskeleton robot with lower limbs is used as the research object, and its dynamic model is established by Lagrange method. Aiming at the serious problem of chattering in traditional synovial control, a fuzzy adaptive synovial control system was designed, and a simulation experiment of trajectory following of unilateral lower limb exoskeleton was carried out. The simulation results show that the sliding mode parameter adjustment mechanism based on fuzzy adaptive can effectively suppress the chattering of the system.

Belt conveyor is one of the main transportation equipment in coal mine. The belt is easy to tear in production. If the tear damage of belt surface cannot be detected in time, it may lead to serious production accidents. In this paper, a belt tear detection method based on industrial camera monitoring is proposed, which can identify the belt tear in time and output the quantitative evaluation result. After filtering the image, Canny edge detection algorithm is used to identify the tear region. A sliding window is used to evaluate the degree of damage area and further determine the control of belt conveyor. Experiments show that the average processing time of a single frame image is 0.4s, which can meet the needs of real-time detection in the production.

Aiming at these issues of complex working conditions and small sample dataset in practical bearing fault diagnosis, a novel method based on improved convolution neural network and transfer learning is proposed. In this method, the global average pooling layer is firstly used to replace the traditional full connection layer as output layer, which can reduce the training parameters of whole network. Meanwhile, the original ReLu activation function is removed and then LeakyReLu activation function is introduced to make the network model be hard to produce Dead ReLu. The improved network is pretrained with the source domain data to learn the fault diagnosis feature. Here, the bottom network parameters are frozen, and the top network is also fine-tuned with a small amount of target domain data to obtain the transfer model. Finally, the target domain data is input to the transfer model for bearing fault diagnosis. The proposed method is verified by considering the scenario of small sample and cross working conditions with the bearing dataset of Case Western Reserve University. The results show that the average classification accuracy of the proposed method can reach 95.49%, and is higher than that of normal convolution neural network.

Based on the coal mine environment is humid, coal mining equipment corrosion is serious, and the existing spray gun short spray distance under the fan spray width is small, spraying distance close to the paint after easy splash, easy pollution after the dispersion of droplet particles and other problems, proposed an air atomization spray gun to its spray paint for corrosion protection, the optimization of its research. The main study of the operating parameters on the liquid surface deposition film formation and airflow expansion degree and other effects. To clarify its spraying performance, Fluent was used to simulate it. To verify the reliability of the simulation results, a spray gun test system was built and compared with the simulation results. Simulation and experimental studies show that increasing the fan control pressure will widen the fan spray amplitude, but if the pressure is too large, the liquid film will be thin in the middle and thick on both sides, and 1.0 bar is preferred. By increasing the pressure of the auxiliary atomizing hole, the fan spraying amplitude decreases, but the spraying quality improves, and 1.3 bar is preferred.

Aiming at the problems of large internal heat generation and uneven heat dissipation of cylindrical power battery module of electric vehicle, a new divergent heat dissipation model of cylindrical battery is proposed. A certain type of 18650 batteries is taken as an example to simulate and analyze its temperature field. The results show that the temperature distribution law of batteries located on the same concentric circle is almost the same. For the single cell, the overall temperature distribution shows the following law: the temperature gradually decreases from inside to outside of the battery. The temperature gradually decreases from top to bottom of the battery . The temperature near the center of the flow field is low. The temperature away from the center of the flow field is high. The change rate of the maximum temperature and the maximum temperature difference of the battery pack gradually decreases with the increase of mass flow rate. The change rate of the maximum temperature and the maximum temperature difference of the battery pack gradually decreases with the increase of the center distance. And the lowest is -1.59% and -14.95%. It provides a reference for the future research on heat dissipation of cylindrical battery

Based on the previous research on the new mandrel type hydraulic buffer, the dynamic characteristics of the buffer are simulated by Fluent software, and the buffer characteristics under different impact speeds are studied. The experimental study was carried out based on the physical prototype, and the experimental results were compared with the simulation results. The results show that the curve trend obtained by the experimental results and the simulation results are roughly the same, which verifies the accuracy of the previous modelling and simulation research. The new mandrel type hydraulic buffer has stable buffer performance, ideal buffer characteristic curve and ideal buffer effect.

In this paper, we designed a new measurement system and applied it to solve the problem of upward sliding of equipment on the heaving face. The LIDAR scanner was used as the data acquisition device in the measurement system, and the B-sample interpolation method was used to fit the surface to the point cloud data, which was very helpful to solve the problem of upward sliding of equipment on the heaving face.

In order to improve the machining quality, improve the machining efficiency and reduce the machining cost of micro-grinding, a nonlinear dynamics micro model was established based on the macro grinding dynamics model, and the nonlinear dynamics equation of micro-grinding was also deduced. The nonlinear dynamic equation was approximated by third order to increase the grinding system parameters. According to the characteristics of the system, the corresponding two-parameter limiting stability lobe diagram is obtained and the stability is predicted. The main parameters include grinding wheel speed, workpiece speed, feed speed, equivalent grinding wheel length and diameter and so on. Finally, the corresponding time domain response graphs and eigenvalue location graphs based on the limit stability graphs of different parameters were drawn to verify the correctness of the stability prediction.

In the process of the airport's facility renovation, construction with suspending air will reduce the airport's operating efficiency, increase the aircraft delay rate, and affect the reputation of the airport. Non-suspend air construction enables airport operations and engineering construction to proceed at the same time, ensuring the efficiency of airport operations. However, there are still some problems in the monitoring and management of non-suspend air construction. Traditional management methods are difficult to ensure the safety of on-site construction and airport operations. With the rapid development of various advanced technologies such as radar detection technology, visible light recognition technology and the Internet of Things, the airport has gradually formed conditions for all-weather and multi-element non- suspend air construction management, multi-source massive data acquisition and intelligent monitoring and management system construction. Through detection radars, night vision cameras, smart wearable devices and multi-source data fusion algorithms, the radar data, visible light data and Internet of Things data are integrated online in real time. Through the electronic fence and visual management platform of the non-suspend air construction area, it is possible to monitor the operation of the construction site, control the positions of construction workers, vehicles and machinery at all times, and prevent the evolution of unsafe events. It can improve the effectiveness of non- suspend air construction safety management, provide a basis for airport management agencies to formulate risk prevention measures, and thereby improve the overall safety level of the airport.

Whether the device can ignite depends on its hot spot temperature, which is closely related to the bridge wire temperature. The 50% ignition excitation of the electric explosive device was determined by means of the up-down method and the statistical theory. The relationship between the temperature of the bridge wire and the injection current was determined by the test. Combined with the 50% ignition current of the electric explosive device, the temperature of the bridge wire was obtained when the electric explosive device is ignited. The accurate ignition temperature provides technical support for the safety performance evaluation of the device.

The paper selected PID Electronic Nose as the representative of gas pollutants’ detection and analysis method, and used it to evaluate the odor of two automotive tape products. The PID Electronic Nose can establish a mathematical relationship based on Weber fischner's law by collecting VOCs concentration and odor subjective intensity of the sample gas. Thus, we can obtain the calculation formula of odor objective intensity of relevant samples. After calculation, the deviation between the odor objective intensity and the odor subjective intensity of the two tape samples is less than 0.3, and the prediction success rate is 100%. It shows that the gas pollutants’ detection and analysis method based on PID Electronic Nose has good accuracy and stability, and can be applied to the gas pollutants’ detection and analysis in vehicle.

Based on the ERA_Interim reanalysis wind data, the best path set data of tropical cyclones from 1993 to 2012 provided by the Meteorological Bureau of India, and the global ocean circulation model ofes data, this paper makes a monthly statistical analysis of the wind, wave, current, tropical cyclones and other marine environmental factors in the North Indian Ocean, providing a reference for manned submersible operation in this area. The results show that the monthly wind speed, significant wave height and average wave period show obvious inter monthly variations. From October to April of the next year, the number of tropical cyclones is less, and the overall sea condition is better. The frequency of level 4 sea condition is less than 16.7%, which is more suitable for underwater vehicles to work on the sea.

Manned submersible is one of the important equipment for human exploration of the ocean. Marine environment factors have an important impact on the safety of its operation at sea. Based on ERA_Interim reanalysis data and OFES model data, this paper takes the Northeast Pacific Ocean as the target sea area and takes Jiaolong manned submersible as an example to make a statistical analysis of the environmental factors such as wind and waves that have a great impact on the submersible operation. The research shows that in the Northeast Pacific Ocean, July-August is the ideal window period for submersible operations at sea. During this period, there are many tropical cyclones and special attention should be paid to weather changes during operations at sea.

Western Sichuan is rich in solar energy resources due to its high altitude, large temperature difference between day and night and low temperature all year round. Vacuum tube collectors have been widely used in high altitude towns in western Sichuan due to their advantages of simple structure, cold resistance and frost resistance. Accurately designing the installation inclination angle of solar collectors is an important condition to improve the utilization rate of solar energy resources. Therefore, this paper selected 6 representative high-altitude towns in western Sichuan to analyse the influence of installation angle on various solar radiation energy collected by vacuum tube collector, the characteristics of the optimal installation angle and the influencing factors of the optimal installation angle. The results show that the optimum installation angle of vacuum tube collector is mainly determined by direct radiation and scattered radiation. For the vacuum tube collector located due south, except for the influence of local latitude, the ratio Ra of scattered radiation to total radiation has a significant effect on the annual optimal inclination angle. In this paper, the optimum installation angle and recommended application range of annual installation angle for vacuum tube collectors in 6 towns are obtained.

An efficient drill pipe delivery system can reduce the time of starting and lowering drilling operations, reduce the labor intensity of workers, and improve the efficiency of drilling construction. At present, most of the ground emergency rescue truck-mounted drilling rigs use manual assistance to change rods, which is less efficient, labor-intensive and has safety hazards. In order to improve the efficiency and safety of rod changing, we developed a hydraulic lifting device which is applicable to various drilling tools and has the function of monitoring the snapping status. The device directly grabs the drill pipe to lift and lower the drill column in the starting and lowering operation, replacing the original way of threaded connection between the output shaft of the power head rotator and the drill column. The buckling monitoring alarm system of the device monitors the buckling status of the lifting device in real time. A finite element method was used to simulate and optimize the structure of the device for two actual operating conditions and key connecting shafts. The results show that strength of the device meets the design requirements. This study should, therefore, be of value to practitioners wishing to drill pipe conveying system.

To improve the efficiency of the submersible pump, a small diameter 25mm pump was developed. At the same time, a laboratory test platform was established to test the performance of the pump and verify the rationality and feasibility of the submersible pump design. The influence of the sinking rate and the liquid level ratio on the submersible pump performance is mainly discussed. The pumping test platform provides test guarantee for the research of efficient pumping equipment, shortens the research cycle and has important influence on field production and efficiency improvement.

This paper presents a bomb simulation model, which includes the simulation of bomb working reliability, the simulation of bomb flight trajectory, the simulation of bomb flying speed and landing time, the simulation of bomb hitting and damaging armored target, the simulation of bomb hitting and damaging process to warfighter, and the encryption of bomb flight path Dimension display) simulation. It is composed of bomb dropping time simulation model, bomb reliability simulation model, bomb flight path and flight path encryption simulation model, bomb landing point simulation model, armored target damage simulation model and kill personnel simulation model. The model not only embodies the main action process of air to ground missile, but also is easy to program.

Aiming at the requirements of motion stability and work efficiency in the actual working process of industrial robots, a time-optimal trajectory planning method for industrial robots with six degrees of freedom is proposed. Taking the six-degree-of-freedom robot PUMA560 as an example, a cubic B-spline curve is used to connect each adjacent path point to express the trajectory, and then the B-spline function trajectory is modified through an improved genetic algorithm to make the robot arm complete the work under the condition of joint constraints with the shortest time.

The object of this work is to investigate the underwater rolling capability of 3-RPS rolling robot. The area transformation matrix is proposed to calculate water flow resistance of the robot under different postures. The rolling ability of the robot under the different water flows is discussed using the ZMP model with the various ground support areas. The initial and end gait of the robot in each support area are planned in order to ensure the roll feasibility of the robot. The roll feasibility study reveals that the greater the water velocity, the more likely the robot will roll when flowing downstream. The robot is in a stable state for the supporting areas A₁A₂A₃ and H₁H₂H₃ , no matter how the water flow speed changes. The robot rolls more easily for the support area D₂D₃F₃F₂ and F₂F₃H₃H₂ compared with the other support areas. The robot can hardly stay and rolls directly for the support area F₂F₃H₃H₂.

In order to improve the performance of permanent magnet synchronous motor (PMSM), an improved index reaching law is presented by introducing the system state variable to solve the problems such as slow tightening speed and high frequency chattering of traditional index reaching law. A fuzzy sliding mode speed control system for PMSM is designed by adjusting the parameters of the reaching law with fuzzy control, and the simulation results are compared with the traditional index reaching law. The results show that the improved approach law can effectively reduce the high frequency chatting at the origin compared with the traditional index approach law, and the PMSM speed control system has faster response speed, better robustness and dynamic and static performance.

In view of the current situation of mining more and more medium thick and steep coal seams in recent years, taking a certain type of shearer drum as the research object, the working state of the drum is analyzed with the help of EDEM software. In this paper, SolidWorks is used to establish the three-dimensional model of the shearer drum. After importing EDEM software, the gangue rock model of the gangue layer with different angles is established according to the position of the 3D model of the drum. The angles of the gangue bedding are 15°, 0° and- 15° respectively. The cutting speed, traction speed and coal seam thickness of the shearer with the same value are adopted, The influence of bedding angle of gangue on shearer drum cutting is studied. The results show that the greater the angle between the gangue layer and the horizontal plane, the greater the impact load on the drum. The working condition of the shearer for cutting medium thick coal seams with gangue is complex and the impact load changes greatly.

Total-environment full-scale accelerated pavement loading facility is a necessary equipment for road testing and evaluation quickly. An ALT-F600 is developed which can realize data collection, analysis and intelligent control of corresponding parameters. The structure and environment unit are introduced in brief. The embedded full environment simulation space is reduced to 52m³, which is more than 40 times smaller than the external workshop environment simulation space, and the energy consumption is reduced by more than 85%.

In order to improve the exoskeleton follow-up performance and avoid the secondary pull injury caused by poor follow-up performance during passive rehabilitation, the passive rehabilitation effect is not good. Accordingly, a robust sliding mode adaptive control law is proposed to improve the following performance. Firstly, the dynamic equation with disturbance is established based on the simplified two-link model of human lower limbs. Secondly, the lower limb rehabilitation exoskeleton control model is established based on robust sliding mode adaptive control law. Finally, the motion control system is built using Simulink, and the main control program is written in Matlab to analyze the tracking performance and adaptive characteristics of the lower limb joints based on the control law for specific trajectories. The simulation results show that the control law has good following performance, strong resistance to external interference, good adaptability, and significantly improves the performance of rehabilitation exoskeleton.

Underwater robot is underwater platforms that integrate movement and perception, and they play an important role in the field of underwater detection. A modular control system is designed on a self-developed underwater robot platform, and its software and hardware control systems were studied in depth. The motion control module, drive module, sensor information acquisition module and image acquisition module are designed in detail. The cooperation of each module effectively improves the control accuracy of the underwater robot. Finally, an underwater test is conducted on the underwater robot, and the results show that the control system can effectively realize the precise control of the underwater robot.

In recent years, medical robots have become a hot spot. Robots assist doctors in spine surgery lamina grinding operations, and the depth of bone grinding on the market depends on the surgeon's hand feeling to perceive the bone layer. Aiming at the perception and recognition of bone layer in the process of lamina grinding, this paper proposes a grinding bone recognition method, constructs the unit energy consumption function of force, feed speed and other signals generated in the process of grinding, extracts its feature vector to represent the bone layer, and then uses the support vector machine method to train and recognize the bone, Finally, the feasibility of this method is verified by a 6-DOF grinding robot. Compared with traditional spine surgery robots, this method effectively improves the recognition accuracy of robot-assisted surgery and ensures the safe and smooth operation of the surgery.

According to the investigation on the use of top coal, the coal seams in different mining areas are different. The number, location, thickness and strength of coal seams are complex and random. The cutting part of drum is the working mechanism of shearer. Its coal rock structure parameters and coal gangue condition will directly affect the working performance of drum. In this paper, the coupling model of drum and coal seam structure is established, and the drum cutting of different coal gangue structures is simulated by EDEM discrete element simulation software. According to the simulation results, the drum cutting characteristics under different working conditions are analyzed qualitatively and quantitatively, and the best drum cutting scheme under different coal gangue structures is obtained.

The object of this work is to design a robot for underwater biological monitoring. First, the overall structure of the robot is designed. The main body adopts a rotating body structure, which middle part is cylindrical and the head and tail are streamlined. The Schatz mechanism is selected for the roaming mechanism of the robot to solve the problem of being easily entangled by various plants and high noise when using propellers. Then, the body of the robot is optimized with the minimum resistance while traveling. The resistance of the robot is discussed when different curve equations are used for the head and tail. Finally, the resistance characteristics of the robot and the surrounding flow field under different inflow speeds are analyzed. The results show that the robot has the least resistance when the Myring curve is selected for the head and the tail with the equation index n=1 and a wrap angle of 20°. In addition, the resistance of the robot and the generated vorticity increase with the increase of the inflow speed, and the shedding of the vortex mainly occurs at the end of the robot's fins.

Aiming at the problems of structure and control of parallel quadruped robot, this paper optimizes the structure of parallel quadruped robots, and analyzes the gait. Firstly, a simplified three-dimensional model is constructed and the force analysis and lightweight processing based on topology optimization carried out for the legs and body parts by Abaqus. Secondly, on the basis of theory, the trajectory of robot foot is reasonably planned and corresponding trajectory equation is deduced. The curve of mechanical leg foot is drawn by using Matlab, and the corresponding angle of motor is obtained by using inverse kinematics analysis. Finally, raspberry PI microcomputer is used as the main control, STM32F407 single chip microcomputer is used as the lower computer, and DJI M3508 motors are used as the driving unit to build the robot experimental model. Based on VMC algorithm, PID algorithm and inverse kinematics solution, the basic motion of quadruped robot diagonal gait is realized, and then the feasibility of mechanical structure and control method is verified.

During the climbing process of the coal mine inspection robot, the driving wheels of the robot are very easy to slip, it is very dangerous. To address this issue, an anti-skid control method based on the fuzzy theory is proposed in this paper. First, the force analysis of the driving wheel in the climbing state is carried out, and then, the driving wheel skid identification method is obtained. On this basis, a skid controller based on the fuzzy control theory is designed to reduce the robot’s slippage. The results of simulation and tests indicate that: the rule setting of the fuzzy controller is reasonable; the controller has strong robustness; the inspection robot based on fuzzy control can effectively restrain the sliding phenomenon of driving wheel and meet the requirements of mine inspection when the robot climbs the slope.

In this paper, a queue batch sampling algorithm is proposed to address the problem of deep reinforcement learning in continuous action control tasks that affects the efficiency of the algorithm due to high sampling correlation. The proposed sampling algorithm reduces the sampling correlation while using the forward value of sampling as a measure to ensure the value of sampling. First, the starting sample is randomly selected as the sampling starting point in the memory bank generated by the robot's interaction with the environment. Second, to save computational resources, the samples in the repository are sampled for the first time, the correlation between the sampling starting point and other samples and the forward multi-step reward of all samples except the sampling starting point is calculated, and the score of each sample based on the ranking position is obtained based on the weighted cumulative results. Finally, the samples with the previous minimum sampling batch size are selected based on the scores to train the deep reinforcement learning model. The proposed queueing batch sampling algorithm is fused with the deep proximal policy optimization algorithm and applied to the forward motion task of a bipedal robot. The proposed improved deep proximal optimization algorithm is compared with the original deep proximal optimization algorithm of the actor-critic framework in simulation, and the learning efficiency of the deep neural network is improved due to the queueing than sampling algorithm which reduces the correlation of sampling and ensures the quality of sampling, and the learning effect of the deep proximal policy optimization algorithm based on queueing batch sampling is significantly improved, that is, the proposed queueing batch sampling algorithm can effectively improve the learning efficiency of bipedal robot learning forward.

In recent years, with the continuous development of artificial intelligence technology, robots began to enter our life. UAV light display, rescue robot and home cleaning robot have brought great convenience to our life, but for more heavy and complex working environment and tasks, the problem-solving ability of single robot is far from enough, and the advantage of multiple robots is particularly obvious. The formation movement of multi-agent system requires multiple mobile robots to move to the target area in the form of formation at the same time, and always maintain a given formation in the process of movement. The group behavior of multi robot cooperation can be controlled by formation control. This paper introduces the research progress and some observation indexes of robot formation control at the present stage, focuses on different formation control methods, and finally looks forward to the development direction of multi robot system in the future. This paper has reference significance for the research of multi robot system.

Combined with the principle of underwater signal transmission, this paper introduces the advantages and disadvantages of various docking and positioning technologies for submarine cable inspection robot to realize underwater charging. On this basis, a multi-sensor fusion docking and positioning technology is proposed, which is applied to the charging and docking guidance of underwater robot, and improves the accuracy and success rate of docking operation.

Laminar decompression surgery is the most common operation in spinal surgery, and it is also considered to be one of the most risky and most difficult operations. In recent years, medical robots have become a hot topic in assisting doctors in laminar flow grinding operations during spinal surgery. In this paper, a spine surgery robot system is established, and the kinematics analysis, workspace and singularity analysis of the 6-DOF surgical robot are carried out. An efficient motion-constrained grinding strategy plan was formulated. In order to verify the feasibility of the strategy, the two grinding plans were compared through the established spinal surgery robot system. The results showed that the efficient grinding strategy has high efficiency and low heat generation. This method can not only reduce the trauma to the patient's bone tissue, but also improve the overall accuracy and efficiency of bone grinding.

A novel redundant actuation 2RPU-2SPR parallel robot is designed. It is a five-axis hybrid machine tool which can be used for complex structure processing in the fields of large ships, new energy and so on. It is composed of parallel mechanism and series mechanism. The combination of the two parts combines the advantages of serial and parallel robots together, enabling the whole machine to have high stiffness, speed, dynamic performance and complex surface treatment capabilities. Next, the forward kinematics was solved, that is, When the parameters of the driving element are known, the parameters of the end-effector are calculated, and the BP neural network optimization strategy is adopted In order to improve the accuracy of BP neural network optimization strategy, the reverse kinematics of the parallel mechanism was derived in detail, and the BP neural network optimization strategy was combined to generate position compensation, so as to achieve high efficiency and high convergence rate. By analyzing the simulation results of parallel robot, it is concluded that this method is suitable for solving the forward kinematics of parallel robot.

Aiming at the stabilization control problem of under-actuated mechanical system, a nonlinear particle swarm optimization algorithm based on sine function inertia weight is proposed to optimize the parameters of PID controller (NPSO-PID) and solve the problem of the under-actuated mechanical system based on the inability to establish an accurate mathematical model. Stabilize the control problem, and use the mobile wheeled robot as the research object to simulate and verify the proposed optimization algorithm. The results show that the proposed nonlinear particle swarm optimization algorithm based on the inertial weight of the sine function has a good dynamic quality and can be improved. The stable control of the under-actuated mechanical system is well guaranteed, simple and easy to implement, and has good practical application value.

To realise the safe and efficient autonomous exploration of mobile robots in unstructured scenarios, this paper proposes an autonomous exploration motion planning method. This method divides the motion planning process into two parts, namely, obstacle avoidance planning for selecting the optimal desired point, and motion state planning for tracking the desired point and generating the motion speed. The optimal and safest target point is obtained by establishing the detection and free planning areas of the mobile robot. Meanwhile, the environment-speed conversion mechanism is established to output the movement speed of the mobile robot. An emergency collision avoidance planning method is proposed to avoid the mobile robot from falling into the local minimum area. Finally, the mobile robot can move safely and steadily in an unstructured scenario. The experiment verifies the safety of the method and the efficiency of exploration. Results show that the proposed autonomous exploration motion planning method for unstructured scenes can realise collision-free exploration and obtain complete scene plane information.

Friction chatter during micro-cutting is one of the negative factors that affect the service life of the cutter and the surface quality of the workpiece. In order to reveal the mechanism of friction-induced self-excited oscillation phenomenon in the micro-cutting system, a micro-cutting dynamic model describes the process of friction chatter between the cutter and the workpiece was established. Two equivalent wear lengths LI and LII were introduced into micro-cutting dynamic model in order to describe the influence of geometric parameter of the micro-cutter and processing parameters on micro-cutting stability. The corresponding non-dimensional amplitude and stability lobe diagram were obtained to change in trend of self-excited chatter. It concluded that two equivalent wear lengths are although different, they give consistent results on the stability of the system. The friction-induced instability can be significantly reduced by adjusting geometric and processing parameters. This provides theoretical guidance for suppressing the frictional self-excited oscillation in the micro-cutting system and improving the stability of the system.

For a 4-wheel steering(4WS) electric wheel vehicle, the control strategy of 4WS and differential steering (4WS-DS) cooperative control is studied in this paper. Based on the idea of hierarchical control, the cooperative control of 4WS-DS is divided into motion control and control distribution layer. In the motion control layer, a 7 degree of freedom (DOF) vehicle model and a sliding mode controller are established to calculate the additional yaw moment required to maintain vehicle stability based on the error between the actual and reference states of the vehicle. In the control distribution layer, a cooperative controller of 4WS-DS is established, and the working areas of 4WS and DS control are divided according to the sideslip angle. Then, the additional yaw moment is optimally distributed to each wheel using weighted least squares. Taking the sine input of ISO7401:2011 standard as the working condition, the simulation by Matlab/Simulink shows that compared with the 4WS and DS control strategy, the sideslip angle is reduced by 68%, and the yaw rate is reduced by 33.3%.

The tendon sheath system is widely used in occasions such as space limitation and long-distance transmission due to its simplicity, flexibility, and remote transmission ability. However, the non-linear characteristics produced by the friction and the flexibility of the tendon reduce the value of engineering applications. And also the change of pretension caused by curvature changing will lead to slack or disconnect of the system. In this paper, based on the previous researches of the tendon sheath, an elastic device connected in series with a double tendon sheath and a friction compensation law with time-varying accumulated bending angle were proposed. The proposed method can significantly reduce the change of pretension caused by bending angle change, with MAPE is 2.34% while the previous method is 25.93% and can significantly improve the anti-interference ability of the system. The proposed method was verified by a simulation and will be applied to the lower limb exoskeleton robots in subsequent research.

In this paper, a three-degree of freedom (DOF) parallel robot is introduced, which can process high-speed difficult and high-precision components in aerospace and automotive engineering. An appropriate reference system is established and the model of the equipment is designed. The kinematics of the mechanism is analyzed, and the complete Jacobian matrix of the mechanism is established by position inverse solution and D-H method. Using the condition number and manipulability as indexes, the dexterity index and the optimum workspace of the parallel robot are evaluated. According to the obtained results, the distribution maps of performance indexes of 3-PUS-UP parallel robot mechanism within workspace are drawn by Matlab software. The research results illustrate that the proposed parallel manipulator has good kinematic dexterity and thus a suitable candidate for five-axis machining.

Taking a rock breaking experimental bench of high pressure water fracturing assisted pick as the research object, the structural dynamic characteristics of the rock breaking experimental bench of high pressure water fracturing assisted pick are analyzed. SolidWorks software is used to establish the three-dimensional model of the experimental bench, and the simplified three-dimensional model is imported into ANSYS Workbench software to establish the finite element model. The modal analysis of the experimental bench is carried out to obtained the first eight order natural frequencies and mode shapes of the experimental bench, analyzed the external excitation frequency of the experimental bench, and find out the parts prone to resonance. On the base of modal analysis, the harmonic response analysis of the experimental bench is carried out by using the modal superposition method, and the amplitude frequency response curves of the key research objects are obtained and analyzed. The amplitude frequency response curves show that the amplitude peaks appear at 31.2hz, 36.0hz, 38.4hz, 57.6hz and 93.6hz. Considering the actual situation, it is summarized that the sensitive excitation frequency close to the first and second order natural frequencies should be avoided in the operation of the experimental bench, which provides a basis for the subsequent optimization design of the experimental bench.

Sampling of seabed rock is an important technical measure for deep-sea resource exploration. Combining underwater bionic technology with ultrasonic drilling technology, a novel robot for submarine rock drilling and sampling is proposed in this paper. The robotic swimming system adopts a bionic jellyfish structure with flexible and stable movement, low noise, outstanding endurance and strong environmental adaptability. The anchoring and drilling device adopts ultrasonic energy to achieve stable adhesion on the seabed and reliable drilling efficiently under low penetrating force and low power consumption. The overall structure and drive control scheme of the robot are presented in this paper. Based on the principle of piezoelectric drive, the ultrasonic anchor and drill are designed. The design of submarine ultrasonic drilling and sampling robot is expected to accelerate the development of submarine exploration technology and promote the effective exploitation and utilization of deep-sea mineral resources.

If the robot wants to improve its work efficiency, it needs to minimize its resistance. A good shell shape can reduce its resistance at a certain extent. For the under water robot designed in this paper, under the condition that the water flow velocity is 0.5m/s and the elevation angle is 0°, the shape of the robot is constructed by three kinds of curves. And three kinds of robot shapes are obtained. These three kinds of curves are ordinary spline, Bezier curve and B-spline curve. The resistances of the three kinds of robots are calculated separately. The result shows that the shape of robot is constructed by B-spline curve design is the resistance least. To further study the B-spline curve, the robot model is constructed with the 3rd B-spline curve, the 4th B-spline curve and the 5th B-spline curve, separately. After calculation and comparison, it is found that the robot constructed by the 3rd B-spline curve is the resistance least, and finally the robot shape is constructed by the 3rd B-spline curve.

A cooperative manipulator was designed. The kinematics model was established by using the Modified-DH method, and the kinematics equation was completed. The Monte Carlo method was used to solve the robotic arm workspace, and the point cloud diagram of the robotic arm end effectors workspace was drawn. The trajectory planning simulation study of the robotic arm was carried out through Matlab, and the displacement, velocity, and acceleration change curves of each joint during the movement were drawn. The rationality of the manipulator structure design was verified, and providing guidance for the subsequent research work of the manipulator.

In this paper, A pipeline robot with variable diameter mechanism is proposed in order to solve the problem of the robot with complicated mechanical structure and a single diameter adaptation problem. The structure composition and working principle of the robot were introduced. The mechanical properties of the variable diameter adaptive mechanism are analyzed and the theoretical model was established. The results show that the pipeline robot has the ability of adapting to the change of diameter and the ability of obstacle surpassing.

By intercepting a limited range of subgrade and imposing an artificial boundary on the boundary of the subgrade, structural designers can complete dynamic machine foundation-subgrade dynamic interaction calculations with existing structural analysis software and finite element analysis software. Based on elastic wave theory, the calculation equations for spring and damper elements of artificial boundary nodes are derived. After proper simplification, only the subgrade parameter is included in the equation for ease of use for designers. Numerical simulation results show that when the subgrade range is 10 times the foundation size, the design accuracy requirements can be met and the calculation efficiency is high.

The monolayer steel frame structure is a kind of the steel structures that has been widely used in the distribution room of Power Substation. Based on characteristics of the distribution room and the design experiences, the designers model and analyse data of the steel consumption with the software YJKS. The paper aims to study an optimization design method and it is economical enough to reduce the steel consumption.

The X5 collapse column encountered in the mining of 30107 working face in Shiquan coal mine was geophysical explored by multi-electrode resistivity method. Combined with the inversion of mine multi-electrode resistivity method, the resistivity response characteristics are analyzed and explained in detail, and the geological structure characteristics of collapse column are obtained. It is found that the decisive factor of water inrush of collapse column is mud section. The geophysical exploration results show that there is a mud section with complete cementation and dense compaction 0 ~ 100m below the trench floor of 30107 belt roadway, which is the water resisting layer of the confined water mining floor of No. 3 coal seam, The existence of mud section ensures the safe mining of the working face. The geophysical exploration results have reference significance for the collapse columns with obvious segmentation of similar geological structures.

In order to promote the waste minimization, the study on microwave in-drum drying technology is systematically researched and demonstrated by using 1:1 engineering prototype. The experimental results show that the average evaporation rate of the drying process is 6L/h ~ 8L/h. The average moisture content of the drying product is less than 5%. The total volume reduction rate is 10%, and the purification coefficient is about 10000. The process indexes meet the requirements of engineering design, and the equipment domestically is completely realized.

Pressure relief by large-diameter Borehole in coal seam is a local method to solve the danger of rock burst. By means of pressure relief drilling in the coal body, So that the pressure relief areas around the drill holes are connected and connected with each other to form a weakening zone, thereby reducing the bearing capacity of the coal seam, redistributing the stress of the coal body under the action of high stress, and reducing the risk of impact of the coal and rock mass, in order to prevent the occurrence of rock burst. A large torque drilling equipment with a diameter of 300mm was developed. Through the field industrial test of Geng Cun Coal mine, through electromagnetic wave CT detection of the coal side of the borehole and comprehensive analysis of the test data, it was found that the proportion of coal and rock mass that can be weakened in the pressure relief zone composed of 300mm boreholes with spacing of 1m was 79.2%. In the pressure relief zone formed by 300mm boreholes with 0.6m spacing, 86.2% of the coal and rock mass can be weakened.

Aiming at the problem that marine organisms entering the water intake threaten the safety of cold source system, a fixed-point acoustic monitoring method for marine organisms in the open water intake channel of nuclear power plant is designed. Ek80 scientific fish detector is used to monitor all kinds of cold source organisms, so as to obtain the density change of juvenile fish and other marine organisms in the open water intake channel in front of the open net interception network, and the efficiency of collecting marine organisms by the open net interception network; The designed acoustic fixed-point monitoring is in front of the first mesh interception network in the seawater open channel, and the beam direction is vertical downward. The monitoring results show that under normal sea conditions: floating marine organisms will enter the interception network through acoustic beam; the larvae of swimming fish with certain swimming ability rarely enter the interception network.

In this paper, the running equipment is detected and the fault is analyzed by the vibration analyzer. The test data shows that there are high 4.94x component peak and harmonic in the spectrum at the pump end, but the impact is not high in the demodulation range spectrum and the bearing is worn in the medium term. Through the analysis and resolution of the bearing fault, the following conclusions can be drawn: doing a good job of daily maintenance can prolong the service life of the bearing, the fault frequency corresponding to the fault of the bearing cage is 0.41x, and the monitoring of the early fault of the bearing can be strengthened.

In this paper, focusing on the problem of dynamic monitoring of existing coal mine metal hoisting derricks, a wireless system for real-time dynamic monitoring of metal derricks based on wireless communication is proposed. On the basis of the original single sensor device, the system monitors the partial force of the metal derrick, the state of the sky wheel, the main body tilt and vibration, etc., through the system integration structure, the various monitored information is gathered together for viewing The data obtained is sent to the AP base station wirelessly through a wireless acquisition device using multiple XLAI analog signal acquisition modules and M series RS485 data acquisition modules, and the AP base station uploads the information to the main control computer for intelligent display, and sets the early warning value for early warning. Field tests show that the design can meet actual needs, can effectively monitor the lifting derrick dynamically, and can make timely and accurate judgments and processing for various fault information, and improve the safety performance and production efficiency of mine operations.

Dyna-Q algorithm has excellent convergence performance and is widely used in planning, control and other fields. Most of the real environment is unstable environment, so it is of great significance to study the performance of Dyna-Q algorithm in unstable environment. In this paper, an environment with controllable instability is constructed and the influence of environmental instability on algorithm performance is studied. Experimental results show that Dyna-Q has better convergence performance than Q-Learning algorithm except in extremely unstable environment, and has the same algorithm validity as Q-Learning algorithm.

In order to improve the accuracy of dynamic parameter identification of SCARA robot, a dynamic parameter identification method derived from dynamic forgetting factor recursive least square method is proposed. Using the Newton-Euler method to establish a robot kinetic model and linearize, select the five-order Fourier grade as an excitation trajectory, and use the genetic algorithm to obtain the optimal excitation trajectory coefficient. On the basis of being extracted, the dynamic adjustment mechanism of the forgetting factor is designed, and the convergence speed of the algorithm is improved. Finally, MATLAB and ADAMS were used to combine simulation experiments, and the minimum inertial parameters of robots were obtained. In order to verify the accuracy of the parameter identification and the validity of the identification algorithm, the trajectory different from the excitation trajectory is randomly selected as input. The reliability of the identification algorithm and the accuracy of the identification algorithm is proved by contrast the synonymous model and the Idams physical model output.

Aiming at the problems of high randomness, many redundant nodes and low efficiency when the map is complex, an improved RRT dual tree algorithm is proposed. Firstly, the algorithm adopts the method of bi-directional growth and adding the target bias factor to the starting point expansion tree to shorten the running time of the algorithm and reduce the probability of exploring the invalid space; secondly, it simplifies the path by greedy thought, removes the redundant nodes, and smoothes the corner by quadratic B-spline curve; finally, through the comparison of simulation experiments. The algorithm has higher search efficiency and better results ,the algorithm has shorter path and higher smoothness, which verifies the accuracy and feasibility of the algorithm.

Job shop scheduling problem is to find the optimal jobs sequence, which minimize the expected makespan. In solving NP-H problems such as job shop scheduling problems by genetic algorithm, trapping in local extremum, low search efficiency and instability are often encountered. In order to restrain this condition, an improved adaptive genetic algorithm based on sigmoid function was put forward. The crossover probability and mutation probability can be adjusted in nonlinear and adaptive based on the dispersion of the fitness of population in the evolution, which is better able to generate new individuals to get rid of the local extremum search to the global optimal solution. To research the performance of the improved adaptive genetic algorithm in solving job shop scheduling problems, a detailed application scheme was given out for the process of it. In the solving scheme, the chromosome decoding algorithm with the objective function of makespan was proposed. Ten JSP benchmark instances were solved with the evaluation index of solution accuracy, convergence efficiency and solution time by the simulation of MATLAB. Through the experimental results of comparing with the other three adaptive methods, the improved adaptive genetic algorithm has been significant improvement in solution accuracy and convergence efficiency.

Path planning algorithm is a key research problem in the application of autonomous mobile robots. RRT algorithm is one of the excellent methods for robot path planning. However, RRT algorithm has drawbacks such as high time consumption, high number of samples, low operational efficiency. On the basis of an adaptive resolution Octree Map, this paper presents an improved RRT path planning algorithm through establishing a growth point evaluation function. By adding the restriction through the evaluation function to random growing points of RRT, the random growth can be turned into purpose-oriented, and redundant growth points and routes can be eliminated. Besides, the growth path is reselected and rewired by the parent node and candidate nodes obtained through the growing process. Through reselection and rewiring, the route is continuously optimized and smoothed. Experiments are carried out for performance evaluation. Experiment results indicate that, comparing with the traditional RRT, RRT* and B-RRT*, the improved RRT algorithm can eliminate the redundant bifurcations of the growth tree reduce the number of sampling times, and greatly improves the growth efficiency.

To overcome the limitations brought about by manual visual inspection in the process of detecting surface defects of metal parts, and to improve the speed and accuracy of detection, a fast detection algorithm for surface defects of metal parts based on YOLOv4-mobilenet Network is proposed. Based on the principle of the MobileNetV3 network, this thesis designs a feature extraction network with faster detection by combining deep separable convolution, inverted a residual structure with linear bottlenecks, and a lightweight attention model based on the Squeeze-and-Excitation(SE) networks. Finally, a deep learning model was trained for surface defects of metal parts. The experimental results prove that the algorithm is faster and the model data volume is smaller than the traditional SSD and YOLOv4 algorithms, which can cope with fast detection tasks.

In order to improve the accuracy of coal gangue detection, this paper proposes to use yolov4 algorithm to detect coal gangue. Through the training and detection experiments of coal gangue data set, it is found that the optimized yolov4 detection algorithm can accurately detect small coal gangue, high-carbon coal gangue, partially masked coal gangue and masked high-carbon coal gangue in different backgrounds, and has considerable real-time, which verifies the feasibility of yolov4 detection of coal gangue.

In this paper, the application of the low-light enhancement method of radiographic inspection is proposed to realize more accurate X-ray nondestructive detection of workpiece defects. In order to solve the problem that the brightness of the original image is too dark in the process of ray image acquisition, gamma correction, histogram equalization, traditional illumination-based method, and deep learning-based method are selected to process the actual workpiece radiographs. Then the advantages and disadvantages of each algorithm are summarized by comparing and evaluating the results of each algorithm, which provides a basis for the selection of low-light image enhancement algorithms in the practical application of radiographic detection.

Aiming at the problem of process route planning, a process route planning method based on directed graph topological sorting is proposed. Firstly, the machining features of the model are identified, and the machining elements are determined according to the feature machining methods and corresponding manufacturing resources. Then, according to the process principles, the directed graph of machining elements is constructed, and all feasible machining element sequences are obtained by improving the topological sorting algorithm. Taking the minimum replacement times of manufacturing resources as the optimization objective, the evaluation function of machining element sequence is established, and the optimal solution is selected as the final process route. Finally, a rotating part is taken as an example to verify the effectiveness and feasibility of the method.

The classic Denisyuk recording method is commonly used to reproduce three-dimensional (3D) image of volume holography, but its diffraction efficiency is low, and white light irradiation is required to obtain high-brightness reproduced image. Aiming at the above problem, the diffraction characteristics of transmissive and reflective volume holographic gratings are analyzed by Kogelnik’s coupled wave theory. A two-step volume holographic recording method is proposed. Firstly, the transmissive volume hologram is recorded and then the reconstructed image is transferred to the reflective volume hologram. Finally, through this method, the brightness and field of view of the reproduced image are improved, and a clear and bright 3D image can be observed under natural light.

The pressure resistance performance of borehole pressure pillows with different thicknesses under working conditions was studied by using the method of combining experiment and simulation. The designed pressure pillow can bear the internal pressure of about 28 MPa and the external pressure of about 17 MPa under the condition of applying the inclusion body. Rupture areas concentrated in the weld area. Based on the theory of elastic-plastic simulation, finite element analysis software ABAQUS was used to simulate the borehole pressure pillow under the condition of the working stress distribution, analyze the inclusion spacing of 45.6 mm and 70 mm case of stress distribution. The purpose of this paper is to determine the type of the operation condition of the working conditions of borehole pressure pillow to ensure the pressure pillow sensitivity.

The work presented in this paper focuses on a novel method that is laser-induced ultrasonic (LIU) imaging detection based on interferometric synthetic aperture focusing technology (InSAFT). The approach adopted is combing interferometry technology with synthetic aperture focusing technology (SAFT). A set of numerical simulations to test the validity of LIUInSAFT imaging detection were performed by the finite element method (FEM). The results of the simulation indicate that the height of target determined by LIU-InSAFT agrees well with that of setting value in the mode of simulation. All the preliminary results throw light on the nature of LIU-InSAFT applied to height detection or inner roughness of inner surface in materials.

Piston pin is an important and key part of engine piston and crankshaft. Its outer diameter accuracy is required to be high. The machining tolerance of 30mm diameter piston pin is 2μm. The accuracy of the testing equipment is less than 0.7μm. Its high precision and surface quality requirements determine that the non-contact on-line detection and measurement method must be used for outer diameter measurement. The changes of temperature, humidity and light in the workshop may affect the measurement results. In order to improve the measurement accuracy, it is necessary to test the relevant factors affecting the measurement system. In this paper, a test-bed is built to test these influencing factors under constant temperature environment and room temperature respectively. The influence of temperature, humidity and light change on the measurement results is analyzed. Among these three influencing factors, the results show that temperature has the greatest influence on the measurement results, humidity has a little influence, and light change (non-indirect sunlight) has little influence on the measurement results.

Kinematic accuracy is one of the most important indexes to evaluate whether the robot has application value and performance. In this paper, the manipulator of 3-PUU parallel robot is taken as the research object. The manipulator has three degrees of freedom. The dynamics and kinematics of the mechanism are analyzed, to deduce the input and output relationship of the mechanism, establish the transfer matrix according to the position error, and then establish the mathematical model related to the kinematic accuracy, and this paper gives the method of accuracy evaluation and calculation method. Monte Carlo method is used to simulate the reliability of the kinematic accuracy of the manipulator structure. Finally, the results show that this method of measuring robot kinematic accuracy is effective, which provides a theoretical basis for the follow-up research on mechanism optimization design and reduce error.

This paper proposes a multi-pedestrian tracking method which can effectively track multiple pedestrians in complex and occluded environments. Pedestrian similarity metrics are calculated utilizing a hierarchical Siamese Neural Network (SNN) to provide similarity scores for candidate pedestrian image regions throughout a time-series of images. We calculate inter-frame pedestrian motion affinity using a difference method. This is merged with the similarity metrics to form a Pedestrian Correlation (PeC) term. The PeC results are processed by the principle of maximum correlation, to achieve pedestrian classification and obtain an identity category for each of several pedestrians. This categorization enables each pedestrian to be associated with different tracklets. A second SNN is then used to calculate similarity metrics for sets of small tracklets in a sliding temporal window. Additionally, tracklet motion affinities, based on mean tracklet velocity, are merged with the tracklet similarity metrics to form a Tracklet Correlation (TrC) term. Finally, the TrC results are also processed by a pedestrian classification method, which categories each tracklet by assigning it to a particular pedestrian. All tracklets for each pedestrian category are then merged to create an overall trajectory for each individual pedestrian. We evaluate our method on two publicly available multi-pedestrian tracking datasets, PETS2009- S2L1 and Town Center, which contain complex environments and pedestrian occlusions. Our method significantly outperforms eight other state-of-the-art trackers from the literature, in terms of tracking accuracy on these datasets.

The research and application of coal-rock interface identification technology is an important part of the intelligent system construction of fully mechanized mining face. Through the automatic identification of coal-rock interface, the cutting height of shearer drum can be adjusted in time to avoid the problems of "under-cutting" and "over-cutting". At present, the research of coal-rock interface identification technology is widely carried out. This paper mainly proposes a coal-rock interface identification method based on ground penetrating radar. Considering the installation and identification accuracy of GPR under actual working conditions, the forward detection method of GPR is proposed. The GPR is installed on the coal mining machine and emits high-frequency electromagnetic waves in front of the coal wall. Due to the different dielectric constants of coal and rock, the characteristics of the received direct waves are also different. By comparing the characteristics, the cutting medium of the current sampling point can be determined, and the interface curve of coal and rock can be obtained by curve fitting.

In order to solve the problem of low heat transfer performance of helical tube under the condition of one-sided heating, twisted tapes in helical tube were built, and the enhanced heat transfer characteristics of the helical tube with twisted tapes were studied through numerical simulation. The result shows: The twisted tapes cause the gas in the helical tube to generate swirling flows in different directions, and divide the gas in the helical tube into different regions, and each region generates a vortex. The twisted tapes increase the gas friction coefficient and Nusselt number in the helical tube, and the helical tube with double twisted tapes has the largest gas friction coefficient and Nusselt number corresponding to the same Re number. Under low Re number, the performance evaluation criteria of the helical tube with double twisted tape are higher than that of the single twisted tape, but when the Re number is greater than 14,500, the performance evaluation criteria of the helical tube with single twisted tape is higher. After the twisted tapes were built in, the gas temperature in the helical tube rises. At low Re number, the gas temperature rise in the helical tube with double twisted tape is the highest. When the Re number is greater than 18500, the gas temperature rise in the helical tube with single twisted tape is the highest.

Safety issues caused by material defects have always been a hot issue that has attracted much attention. How to realize rapid and accurate identification of material defects is the focus of research on material defects today. Barkhausen noise measurement is a new material performance testing technology, which is suitable for ferromagnetic materials and is sensitive to many characteristics of the material. This paper builds a Barkhausen inspection system to detect and quantitatively analyze the defects of different widths and depths in ferromagnetic materials and explore the influence of different excitation parameters of the inspection system on the Barkhausen signal.

With the rapid development of artificial intelligence technology, improving the identification ability of the machine and making the machine have "senses" ^[1] will further improve the working efficiency of the robot in both production and life. In the first mock exam, this paper focuses on the classification of objects under the fusion of vision and tactile features. Taking common objects as the research object, the visual and tactile information data of objects are collected. The characteristics of target objects are extracted by using image processing algorithm and tactile sensor signal principle. The redundant features are removed by using two dimensionality reduction methods: multi cluster feature selection algorithm (MCFS) and principal component analysis algorithm (PCA), and the effective features are put into support vector machine Library (SVM), two different feature dimensionality reduction methods are fused through decision level fusion to optimize the classification accuracy.

In order to solve the problem of uneven airflow distribution of Laval nozzle in jet crushing, a comparative analysis of supersonic jet performance in nozzles with multiple contraction curves was conducted. The study found that the contraction section curve not only changes the velocity and pressure distribution of the contraction section of the nozzle, but also affects the uniformity of the flow in the nozzle, and the uniformity of the throat airflow and the contraction curve of the "slow flow" region has a certain relationship, the longer the region, the better the uniformity of the throat airflow; and the contraction curve caused by the airflow in the nozzle velocity, pressure and uniformity differences caused by the contraction curve in the nozzle with the supersonic flow of gas in the expansion section is gradually weakened.

In recent years, with the continuous development of science and technology, the intelligent level of coal mining requirements continue to improve, one of the key points is the up and down of fully mechanized mining working face equipment, for the current fully mechanized mining working face straightness measurement error is large, the sensor anti-interference ability is poor and other problems, In this paper, an on-line measurement system for up-sliding migration of fully mechanized mining face equipment based on lidar is presented. The application of lidar sensors improves the angular resolution, ranging accuracy and anti-jamming ability of the system. The system uses ARM processor to improve the execution speed and performance of the system.

As a key rotary connection component of construction machinery, the operation performance of slewing bearing has an impact on the stability of engineering construction. Condition monitoring for slewing bearing is essential to their high availability and profitable operation. However, the characteristics of slow-speed large-size slewing bearing make the weak vibration signal corrupted with noise. Therefore, effective signal de-noising for preprocessing technique is difficult but crucial. To solve this problem, a novel signal de-nosing method using robust local mean decomposition is proposed with a product function selection strategy based on kernel principal component analysis. The effectiveness is validated by using simulated as well as experimental vibration signals obtained through a slewing bearing highly accelerated life test. The results illustrate that proposed method can perform effective signal de-noising of slewing bearing compared with other conventional method.

The stress of weld in pressure vessel is complex and the working conditions are poor. The conventional nondestructive testing methods cannot meet the requirements of weld surface defect detection. In this paper, eddy current pulse thermal imaging method is used to study the detection technology of weld surface defects. The finite element simulation software is used to analyze the electromagnetic induction eddy current and temperature distribution at the weld crack. Build an experimental platform to verify the feasibility of eddy current pulse thermal imaging to detect weld cracks. In order to eliminate the background noise of the thermal image, the detection result adopts the absolute temperature rise method to improve the detection accuracy.

Aiming at the problems of difficulty in positioning without GPS signals in the mine scene, the traditional wireless positioning method has a long deployment time and rather low accuracy. The 3D lidar point cloud technology is selected for localization, based on the 3D point cloud SLAM technology. Because the highly similar geometric characteristics of the passageways of mines, the indoor corridor environment was chose for simulation experiments. First, a priori three-dimensional point cloud map was established, then the current frame point cloud matched with the prior map, and finally the pose in the prior map obtained from the matching to be used for tracking. A high-precision dead reckoning odometry was used as a priori pose and redundant localization to increase the system robustness when the lidar odometry fails. For relocation with high real-time and accuracy requirements, NDT(Normal Distributions Transform) matching algorithm was used with two deficiencies of the are improved. First, grids was used to divide the point cloud into different clusters, and then the common size of clusters was used to dynamically determine the NDT resolution; Secondly, the speed of ikdtree and kdtree at rebuilding the tree and searching for nearest neighbors was evaluate, and finally the ikdtree was choosen to replace kdtree to acceleration the process of NDT matching algorithm. And the algorithm has tested at our intelligent car platform.

In order to study the operation of planetary gear transmission, the article takes the two-stage planetary gear transmission system as the research object to determine the correct meshing parameters between the sun gear, planetary gears and internal gears. With the help of a plug-in in SolidWorks, a three-dimensional planetary gear is carried out. Modeling and virtual assembly, using ANSYS Workbench to simulate and analyze the planetary gear transmission. Through the direct integration method of transient dynamics, the equivalent stress, equivalent strain, and equivalent deformation of the planetary gear tooth meshing are obtained, which can provide reference for the future planetary gear transmission design and simulation analysis.

This paper presents the consensus control with diverse time-delays on multi-agent systems. Depending on the comparison among the performances (the speed of convergence, robustness, and synchronization) of different graphs on the consensus problem without the mathematical analysis of algebraic connectivity, the current study aims to emphasize the importance of the structure of networks, comprising nodes and edges, in control engineering. This research is based on three typical complex networks (Balanced digraph, Regular Lattice-Type Nearest-Neighbor Graph, and Small-World Network) whose initial state, number of nodes (N_V = 10), and edges (N_E = 20) are consistent. The most significant difference among networks is the degree of graphs. The simulation results of the consensus control with different time-delays on these graphs with single integrator dynamics agents are collected to find the relationship between the degree of network and time-delays. The communication time delays are chosen from τ = [0,0.1,0.2,0.23,0.3,0.4]. Finally, it is found that graphs with bigger degree have better robustness to time-delays. And there is a general trade-off between convergence speed and robustness to time-delay, though slight time-delay may benefit the consensus. What’s more, it is proved in this study that the degree of graphs is not sufficient to determine the accurate maximum of time-delays.

Aiming at the wear state problem of heading machine, a wear state prediction model based on Wiener process was established. The vibration and acoustic emission data samples were extracted by setting up a test system, and WPD was applied to denoise the data. Six kinds of pick wear degree states were defined, and 50 groups of data samples were taken from each state to verify the accuracy of the model, which all met the accuracy requirements. Then, the model was used for data prediction research, and the experimental data were compared. The results show that the prediction accuracy of vibration signal acceleration energy and lower Wiener process is 99.01%, and that of acoustic emission signal acceleration energy and lower wiener process is 99.13%.The prediction error is small and the prediction accuracy is high, which provides a new method for the prediction of heading machine pick wear degradation state.

As an important water conveyance structure, aqueduct plays an important role in the development of national economy. It is very necessary to study the seismic performance of aqueduct. Taking an aqueduct in Northwest China as the research object, considering the vulnerability of the basic components of the aqueduct, the vulnerability of the overall structure of the aqueduct is established through the joint function, and the seismic performance of the basic components of the aqueduct and the overall aqueduct is analyzed.

At present, the communication technology of underground coal mine mishgas safety monitoring system mainly adopts a single communication mode, which has problems such as network interruption and network delay, thus causing the communication delay or communication interruption of safety monitoring system, resulting in coal mine accidents. To solve this problem, a multi-path optimal communication technology (MPOC) composed of multiple communication paths is proposed, which can effectively manage the communication paths of safety monitoring system, optimize the communication time of the system, and make the system communication in the optimal communication path. Many field experiment data proves that MPOC technology can ensure the optimal communication path of safety monitoring system and normal operation of the system, avoid coal mine accidents caused by the failure of monitoring system and ensure the safety of coal mine production.

Coal plays a dominant role in China's energy supply, so coal mine safety production is particularly important. Mine fire is one of the common coal mine disasters, and the goaf is the most prone place for underground fire. Therefore, the prediction and prediction of goaf fire is very important for mine fire control. Strengthen the research on goaf fire prediction and prediction, It is a great guarantee for the safety of underground workers. In this paper, taking the goaf of 3117 working face of Xin'an Coal Mine as a reference, the beam tube monitoring system is used to analyze the extraction of index gas in the goaf, and the three spontaneous combustion zones are divided according to the oxygen concentration. Finally, the minimum advancing speed of the working face is calculated to realize the prediction of spontaneous combustion fire in the goaf of the working face.

China is rich in coal resources, and its reserves and consumption rank among the top in the world. Mine fires and coalfield fires caused by coal seam spontaneous combustion have brought great difficulties to coal mining and environmental governance. In this paper, the infrared thermal imager is selected for the field test in the high-temperature abnormal area of coal seam spontaneous combustion. Combined with the correlation and linear relationship among the temperature, pixel value and distance of the infrared thermal imager, the field test data and temperature fitting curve of Daquan Lake fire area in Xinjiang are obtained. The test images at different distances show that the closer the test distance is, the higher the pixel value of the instrument is, and the clearer the fire zone is; the fitted maximum temperature curve shows that the closer the test distance is, the higher the maximum temperature is, and the linear relationship is obvious; according to the attenuation law of the pixel value of the infrared thermal imager with the increase of distance, it can be seen that the maximum temperature of the test image has a linear relationship with the test distance. The closer the test distance is, the higher the maximum temperature is. This study confirms the qualitative and quantitative law of infrared detection in the high temperature abnormal area of coal seam spontaneous combustion, and provides a certain reference for the research of infrared detection of coal seam spontaneous combustion.

As an indispensable tool for people to travel, elevator plays an important role in our daily life. Through investigation and data analysis, this paper deeply studies the faults of elevators in use in Shanghai in recent three years. This paper analyses the fault systems and fault types of different types of elevators, further studies the causes of faults, and puts forward some suggestions on how to reduce the number of elevator faults and improve elevator safety.

In order to make clear the cutting performance of spiral drum in coal and rock strata with small faults, this paper takes MG500/1130-WD shearer as the research object, uses SolidWorks to establish three-dimensional model of spiral drum, and uses EDEM to establish angle of 30°, 35° and 40° coal wall with coal seam angle, and to simulate the process of cutting coal wall with fault. The cutting resistance and rock quality in the simulation process are extracted, and the cutting resistance force, productivity and specific energy consumption of the front drum and the rear drum in different stages of crossing fault are analyzed in detail. The results show that: (1) with the increase of coal seam angle, the cutting resistance of spiral drum increases, and through the cutting resistance load fluctuation coefficient to explore the more reliable test conditions of this type of shearer. (2) With the increase of coal seam angle, the productivity of shearer shows a downward trend, and the cutting specific energy consumption increases with the increase of coal seam angle. The influence of coal seam angle on the cutting specific energy consumption is verified by single factor variance method. This paper provides a certain theoretical support for the shearer spiral drum to cut coal and rock with faults and select cutting parameters.