Abstract:A new type of variable friction damper is proposed and designed to improve the energy dissipation capacity and stability of the traditional friction damper. The new variable friction damper can achieve variable damping force output characteristics and have a certain anti-eccentric load capacity by adopting a reverse discontinuous slope structure design and a butterfly spring combination, and it can effectively improve the energy dissipation capacity of the damper, because the auxiliary friction plate is also added to the structure. Firstly, the mechanical characteristics of the damper in different working stages are analyzed by establishing a theoretical model, and the influence of various design parameters on the damping output characteristics of the damper is analyzed. Based on this, a prototype of the processing principle was designed and the mechanical characteristics were experimentally studied. Finally, the anti-eccentricity of the isolator was verified by the finite element method. The results show that the test results of the new variable friction damper are consistent with the theoretical analysis；the hysteresis curve is “dog-bone”，that is，the output characteristic of variable damping force can be achieved；compared with ordinary flat friction dampers，the frictional energy consumption can greatly improve the energy consumption capacity. Different damping characteristic outputs can be achieved to obtain different vibration isolation requirements by adjusting the value of the torque. The analysis results show that the new variable friction damper has a certain resistance to eccentric load.

Abstract:To investigate the effect of the external stationary shoulder on the formability and mechanical properties of friction stir welding joint，the self-developed NRSA-FSW ( non-rotational shoulder assisted friction stir welding ) was used to weld 2219-T6 aluminum alloy with the thickness of 4 mm plate. Mechanical properties of the joint were analyzed by tensile test and microhardness test. Optical microscopy and field emission scanning electron microscope were used to analyze the microstructure of the joint cross section. The results showed that the surface of the joint processed by NRSA-FSW formed without fish scales，flash，and thinning，and represented smooth appearance due to the scrapping effect of the static shoulder. Compared with the joint processed by FSW (friction stir welding) under the same condition，the softening zone of the joint processed by NRAS-FSW increased，and the microhardness distribution of the joint was more uniform. The average ultimate tensile strength of joints processed by NRAS-FSW was 323 MPa，approaching 72% of the base material，which was similar to the joint processed by FSW. Defects were not observed in the joint processed by NRSA-FSW. NZ (nuggest-zone) appeared as a "U" shape and the joint fractured in the NZ during tensile tests，which was ductile fracture.

Abstract:In order to overcome the problem that the existing urban road lighting street lamp control systems are not able to balance energy saving and driving safety，an adaptive energy saving control algorithm for urban road lighting is proposed and studied. The algorithm uses real-time average brightness feedback on the road surface，compares the brightness of the road surface based on the standard design during sunrise and sunset hours，adaptively adjusts the actual switch light time. According to the average traffic flow of the road hours，the corresponding road lighting brightness is calculated by the hour during the night lighting period，realizing the time-division dimming control strategy that matches the statistical results of the traffic flow，which can guarantee driving safety and realize obvious energy saving. It describes the working principle of the algorithm and the flow chart of the program in detail，and introduces the structure and function of the urban road lighting monitoring system based on the LoRa wireless internet of things. The actual operation effect of the system in the same weather conditions and time on a main road in Hunan Province in summer shows that the system can reliably realize the monitoring functions of the remote wireless switch and dimming control of the regional road street lights，verifying the effectiveness of the energy-saving adaptive control algorithm，with an energy saving rate of 39.3% compared with the traditional control methods.

Abstract:In order to explore the feasibility of magnetorheological elastomer (MRE) used in variable stiffness semi-active vibration isolation，silicone rubber based MRE was prepared; the MRE isolator was designed and manufactured; the MRE material and the performance of the isolator were tested; the variable stiffness semi-active vibration isolation system based on MRE isolator was built，and the on-off control，fuzzy control and GA fuzzy control algorithms were designed. The test results show that the storage modulus of MRE reaches saturation when the magnetic induction intensity is 600 mT，and the magnetorheological effect reaches more than 2 000%; the variable stiffness characteristic of MRE isolator is obvious，when the coil current changes from 3 A to -3 A，and the equivalent stiffness increases from 74.216 kN/m to 137.128 kN/m with an increase of 84.8%. After optimization，the GA fuzzy control effect is improved by more than 10%，and the control effect of the three methods on the peak value of displacement response is more than 50%，and the control effect on the RMS value of displacement is more than 60%. The results show that the effect of GA fuzzy control is obviously improved after optimization，and the magnetorheological elastomer vibration isolator can be effectively applied to the semi-active vibration isolation control system with variable stiffness.

Abstract:Acoustic Emission（AE） monitoring of hard and brittle and hard ground zirconia and alumina engineering ceramics created in their ground surface was studied，and the correlation between the removal volume of a single diamond abrasive scratch ceramic materials and its corresponding acoustic emission signals were studied. The experimental results show that with the increase of the removal volume of a single diamond abrasive scratched ceramic，the signal energy ratio of the acoustic emission signal in the low-frequency range increases，and the eigenvalues such as root mean square，maximum amplitude and standard deviation of acoustic emission signal increase，and the increasing trends of the two ceramic materials are different; But the characteristic values of these acoustic emission signals are all related to the corresponding removal volumes of the two kinds of ceramic materials. At the same time，time series autoregressive AR（2） model of the acoustic emission signals of a single abrasive scratch ceramics can completely represent the corresponding acoustic emission signals. The absolute values of the time series autoregressive AR (2) model eigenvalues increase with the increase of the two kinds of ceramics scratch removal volumes，and the increasing trends are similar to that of the acoustic emission signal eigenvalues. Acoustic emission signals can be an ideal data source for the online monitoring of surface creation in grinding processes of engineering ceramics.

Abstract:Aiming at the parallel parking path planning and tracking control problem for the wire-four-wheel steering vehicle，a novel parallel parking path planning method is proposed based on the improved particle swarm optimization algorithm，and a novel parallel parking path tracking control strategy is proposed based on feedforward and feedback control architecture. Firstly，considering the kinematic non-holonomic constraint of the wire-four-wheel steering vehicle，the process and boundary constraints of the power and steering sub-systems，the obstacle avoidance constraint，the initial parking pose constraint and the target pose constraint，a parallel parking path planning constraint optimization problem with minimizing the parking time is established，and the particle swarm optimization，which can deal with the equality and inequality constraints，is adopted to solve the constraint optimization problem. In order to execute the planned parking path，the desired wheel steering angle is used as the feedforward control quantity，and the PI feedback control quantity is constructed based on the deviation between the actual pose and the expected pose of the vehicle. Finally，the model-in-the-loop simulation system is built based on vehicle dynamics simulation software to verify the feasibility and effectiveness of the proposed method，and the simulation results show that the proposed method can guide the wire-four-wheel steering vehicle to complete the parallel parking task quickly，accurately and stably.

Abstract:Aiming at the difficulty of heat dissipation during long-term operation of in-wheel motor electric vehicles，a matching cooling mechanism and cooling method are designed. On this basis，a thermal and magnetic coupling temperature field simulation model of the in-wheel motor is established. Under the two operating conditions，the thermal and magnetic coupling temperature rise of in-wheel motors based on different cooling sources and multi-mode modes are carried out，and the simulation results are verified by experiments. The results show that the motor high-temperature problem is significantly improved during the long-term operation of electric vehicles; the two operating conditions have different effects on the motor temperature rise. The simulation results are in good agreement with the test results. The maximum winding error in working condition 1 is 5.1%，the maximum rotor error is 4.9%，the maximum winding error in working condition 2 is 4.8%，and the maximum error of the rotor is 4.5%. The study of high-temperature problems during long-term automobile operation provides a certain reference.

Abstract:To effectively deal with structural optimization problems with epistemic uncertainty，an evidence-theory-based reliability design optimization method using approximate moving vectors is proposed. It first converts the evidence variables into probability variables and constructs an equivalent probabilistic reliability-based design optimization model. Through solving this model using the sequential optimization and reliability assessment method，an approximate design point is obtained. Then，the evidence-theory-based reliability analysis is carried out for each constraint at the design point，based on which the approximate shifting vector and deterministic optimization model are established. A new design point is obtained by solving the deterministic optimization problem. Finally，the sequential iteration process composed of equivalent probabilistic reliability-based design optimization and evidence-theory-based reliability analysis is repeated until convergence，and the optimal design point is obtained. The proposed method can convert the nested evidence-theory-based design optimization problem into an iterative solution process，which can effectively reduce its computational cost. The effectiveness of the proposed method is verified by three examples.

Abstract:This work analyzed the influence of the position of relay coil on the output power and transmission efficiency in the three-coil wireless power transmission system. A transmission strategy using a DC-DC conversion circuit was designed to maintain the transmission efficiency at a high level by adjusting equivalent impedance of load without changing the placement of the relay coil. The experiments show that the proposed transmission strategy is not only suitable for the situation where the relay coil is fixed and load resistance is changed，but also can maintain high transmission efficiency when the load remains the same but the position of the relay coil changes.

Abstract:In order to solve the problem of imperfect line protection in medium voltage DC distribution network，a single-pole fault protection method based on zero-mode power is proposed in this paper. Firstly，the characteristics of single-pole fault in the DC distribution network are analyzed. Fault identification is carried out by using the polarity of the transient current component，fault line selection is realized by using the magnitude of zero -mode power amplitude，and fault pole selection is implemented by using the magnitude of zero -mode voltage. Secondly，the setting criteria are given for the proposed protection method. Finally，a DC distribution network model is built in PSCAD/EMTDC to output fault data，and MATLAB is used to verify the protection method. The simulation results show that the proposed method is fast and effective，highly reliable，and has a strong ability to withstand transition resistance. It is not sensitive to data delay and has a strong anti-interference ability.

Abstract:In order to solve the problem of accurate measurement on the parameters of low-voltage power supply lines，in this paper an intelligent measurement method of low-voltage power supply line impedance based on phase-to-mode conversion is proposed. And the least-squares optimization model is established to optimize the impedance parameters to achieve accurate measurement. Finally，the practical example verifies that the method is effective and feasible. The method realizes the online identification of the impedance parameters of the low-voltage power supply line under different working conditions，and provides a basis for the evaluation of the operation state of the power supply line.

Abstract:To solve the problems of stability and load uncertainty of Buck DC-DC converters with constant power load (CPL)，a model predictive control（MPC） strategy with high-order sliding mode observers is proposed. Firstly，the objective function of the Buck converter is constructed according to MPC theory，and the rolling optimization tracking equation is established to solve the optimal control rate. Secondly，a high-order sliding mode observer is constructed to improve the voltage control accuracy and eliminate chattering and the relative order problem. Finally，the stability analysis of small disturbance is carried out to establish the equivalent model of source and load side. When CPL and resistive load are switched frequently，the simulation results show that the MPC control based on high-order sliding mode observer has good dynamics and robustness，compared with the traditional double closed-loop PI adjustment and MPC control. In addition，it can control and track the bus voltage when the load changes.

Abstract:To solve the loss of zero-voltage switching (ZVS) performance of traditional DC/DC phase-shifted full-bridge converters under light load and the serious reverse recovery problems of the secondary diodes，in this paper，by moving the filter inductor on the secondary side of the traditional DC/DC phase-shifted full-bridge converter to the primary side，the zero-voltage switching (ZVS) performance of all the switches in the primary-side under heavy load is achieved，and the zero-current switching (ZCS) performance of switches in the lagging bridge can be achieved under light load. In addition，the current of the secondary-side diodes obtains natural zero-crossing characteristics under heavy load and enters in discontinuous conduction mode (DCM) under light load，which effectively weakens the diode reverse recovery problem. Continuous conduction mode (CCM) and DCM of the topology are described in detail，and the realization mechanism of ZV-ZCS characteristics is clarified. By establishing the gain equations in different modes of the topology，the design rules of key parameters are inferred. Finally，simulation based on PSIM software and a 2 kW experiment prototype are developed to verify the characteristic of the topology and the correctness of the theoretical analysis.

Abstract:In this paper，The design of a low side-lobe compressed dipole resonating at the third mode is proposed in order to solve the problem of a high side-lobe in the mode compressed dipole antenna. This paper adopts a method of loading a bent inductor on the dipole antenna to achieve a larger compression coefficient，thus the radiation interface of the reverse current is reduced. Finally，the target of designing mode compression antenna with low side-lobe and high gain is realized，which achieves a low side-lobe of -15.35 dB and a gain of 5.11 dBi. The mutual interference is reduced，and the transmission distance and speed are increased. In order to verify the validity of the simulation model，the PCB technology is used to manufacture the model and test it. The simulation results are in good agreement with the test results.

Abstract:The second-order temperature compensation was realized by weighted superposition of the first-order temperature compensation voltage and the voltage with high-order temperature term which was conversed from the subthreshold leakage current of MOS transistor. In addition，a high-gain operational amplifier and negative feedback loop were adopted to improve the power supply rejection ratio（PSRR）. Subsequently，a bandgap voltage reference with low temperature drift and high power supply voltage rejection ratio was proposed. Based on 0.18 μm CMOS technology，circuit design and simulation，layout design，and post-simulation were carried out. The results indicated that the output voltage was 1.22 V under the power supply voltage of 1.8 V；the temperature coefficient（TC） was 3.3 ppm/℃ in the temperature range from -40 ℃ to 110 ℃；the PSRR at low frequency was -96 dB@100 Hz；the static current was only 33 μA.

Abstract:A Σ-Δ ADC modulator with low voltage and low power consumption is proposed for the application of Σ-Δ ADC in audio equipment. The modulator structure improves the traditional modulator structure and optimizes the modulator coefficients，thus overcoming the shortcomings of the traditional Σ-Δ ADC modulator structure. At the same time，OTA amplifier and comparator，two key circuits in the modulator，are optimized，which greatly improves the performance of OTA amplifier and comparator. The improved modulator has the characteristics of low voltage，low power consumption，high precision and good robustness. The modulator is powered by a 1.2 V low voltage，and it has an over-sampling ratio (OSR) of 128，a sampling frequency of 6.144 MHz，and a signal bandwidth of 20 kHz. Based on SMIC0.11 μm process，the layout design of Σ-Δ ADC modulator was completed，and the streaming chip was finally successful. The measurement results after chip streaming show that the signal-to-noise ratio of the modulator reaches 102.4 dB，the effective bit reaches 16.7 bit，the overall power consumption of the modulator is only about 1.17 mW，and the whole layout area of the modulator is only about 0.122 mm2. The measured performance of the modulator shows that it is a good choice for Σ-Δ ADC circuit in audio chip.

Abstract:In order to improve the accuracy of bearing fault feature extraction and operation condition evaluation，a diagnosis method based on wavelet packet dispersion entropy and Meanshift probability density estimation is proposed. Firstly，wavelet packet transform is used to increase the dimension of bearing vibration signal data，and the dispersion entropy (DE) of each sub-band is calculated to construct the characteristic matrix. Then，PCA is used to reduce the dimension of multi-dimensional matrix visually. Meanshift nonparametric estimation is used to obtain the maximum probability density position of training samples as the clustering center. Finally，the Euclidean distance between the test sample distribution entropy coordinates and each cluster center is calculated to determine the test sample category. The experimental data of CWRU and QPZZ-II are used to verify the effectiveness of the proposed method for identifying different fault types and fault degrees. Due to the complete theoretical model of wavelet packet and the ability of signal band decomposition sparsity，combined with the good robustness of the DE index，the constructed feature matrix has good aggregation and large inter-class distance. At the same time，Meanshift aims at maximizing probability density，and can effectively classify different data samples by adaptive iterative clustering center and membership.

Abstract:Aiming at the problems of a long time and poor repair effect of the current surface image information restoration of automotive strip steel，a method for repairing automotive strip surface image information based on sparse representation is proposed. According to the spatial correlation information existing between the pixel points in the automotive strip surface image，combined with the gray distribution information of the pixel points，the two-dimensional histogram is used to establish the dispersion measure matrix，and the target area and background area needed to be repaired in the surface image of the automotive strip steel are obtained. The sparse repair model is constructed based on the sparse representation principle，and the damaged area is repaired in the target area and background area of the automotive strip surface image，which can effectively obtain the image information of the automotive strip surface defect and realize the automotive strip repair of surface image information. The experimental results show that the proposed method has a high peak signal-to-noise ratio and small root mean square error for the image information restoration of the automotive strip steel surface，indicating that the method has a better information restoration effect，the restoration time is shorter，and the restoration efficiency is higher.

Abstract:To solve the problems such as over-reliance on massive data and weak prior generalization ability in the training procedure of image super-resolution，thus further to improve the quality of reconstructed high resolution image，a new image super-resolution algorithm was proposed. This paper firstly extends the theory of image self-similarity and points out that the self-similarity of ideal reconstruction image is extremely strong，but this property can be sharply weakened when the reconstructed image is attacked with some degradation factors. Then this discovery is considered as a prior and described by constructing a joint Gaussian mixture model，so that the self-similarity of each reconstructed image patch in the prior term can be represented by a specific Gaussian distribution. For maintaining the training samples' consistency，only the image patches extracted in the input image closed to its spatial position are permitted to join in the modeling process for each high-resolution image patch. This style can avoid the step of finding the nearest neighbors which is liable to introduce errors. Finally，the whole high-resolution image can be reconstructed patch-wise in an iterative way. Extensive experiments demonstrate that the reconstructed images generated by the proposed algorithm are clear and natural，in which the salient edges and texture structures are effectively preserved，and the correct high-frequency information is recovered. The 3× super-resolution experiment in BSD500 shows that the average PSNR is higher 0.529 db than the state-of-the-art algorithm MMPM，and the average SSIM is 0.030 higher than MMPM.

Abstract:In order to achieve precise picking of residual film in farmland and to improve the recovery rate of residual film recovery machine，the improved Faster R-CNN convolutional neural network is applied to identification and detection of residual film in farmland，and a method of identifying residual film in farmland is proposed. Taking the residual film left on the surface of the farmland after the 11MS-1850 residual film recovery machine worked as the research object，a total of 1 648 images are collected during different periods of sunny and cloudy days. The data set is expanded by changing the image brightness，rotation，etc，and finally 4 950 residual film sample images are got，which are divided into a training set (3 465)，a validation set（990），and a test set (495) according to 7 ∶ 2 ∶ 1. The dual-threshold algorithm is used to replace the traditional single-threshold algorithm，which reduces the impact of thresholds on model performance. Through comparative experiments，ResNet50 with a residual network structure is selected as the backbone feature extraction network. The accuracy rate can reach 88.84%，the recall rate is 87.70%，and the overall accuracy is 88.27%. In order to make the detection model more sensitive to small targets，according to the size of the residual film in the data set，the scale parameters of 322 and 642 are added to the original anchor points，and the accuracy，recall，and overall accuracy are improved by 1.29%，0.67%，0.97%，respectively; the single detection time is 284.13 ms，which basically meets the requirements for identifying residual film. It can provide a reference for the installation of replenishment equipment for the residual film recovery machine，and provide a theoretical basis for the development of artificial intelligence residual film recovery machines.