Abstract:High-precision permanent magnet synchronous motors have extremely high requirements on the dynamic performance and speed stability of the speed control system. Aiming at the problems of poor dynamic performance of traditional motor control methods and large interference from uncertain loads, a speed-current non-cascaded direct speed controller was designed using the method of model prediction, which replaced the traditional proportional-integral (PI) regulator. Eliminate the double-ring cascade structure. The prediction model performs long-distance prediction on the speed, and uses a more restrictive cost function to optimize the dynamic and static performance of the control system. Finally, the simulation experiment shows that compared with the traditional PI vector control, the speed overshoot is reduced to zero, the maximum speed drop is reduced by 45r/min when the load is suddenly applied, and the recovery time is accelerated by 0.055s, at the same time, the effect of suppressing the fluctuation of speed and current is significant, and the system dynamic performance and anti-interference ability are better.

Abstract:The peak-to-valley current control Buck converter has symmetry for the influence on the stability of the system when its own parameters change. Taking the Buck converter as an example, a symmetric piecewise linear model of peak-to-valley current control is established, and the corresponding Jacobi matrix is derived. Based on this, the symmetrical threshold of the output capacitor time constant is derived when the system is in stable state transition to unstable operation. Finally, the experimental results are verified by constructing the simulation circuit in PSIM software, the results show that the peak-to-valley current control Buck converter has symmetry in the working state of the system when the parameters change, and the output capacitor time constant threshold value proposed in this paper is very important for the system parameter design and device selection.

Abstract:The traditional water temperature regulation adopts manual experience and semi-electrical automation. These methods have the problems of single structure, large water temperature fluctuation and poor stability. Therefore, a PID correction control system based on increment is designed. Based on incremental PID correction control system, thermal resistance is used as temperature sensitive element, and a three-wire thermal resistance temperature data acquisition bridge circuit is designed. According to the experimental temperature data and its thermal resistance, the correction coefficient is calculated to reduce the data output error. Finally, the experimental prototype is designed and made to verify the correctness of the theoretical analysis according to the data obtained from the prototype.

Abstract:Time-interleaved Analog-to-digital Converter（TIADC）is an effective way to increase the sampling frequency of ADC, but the sampling time and relative error between the channels seriously affect the system performance. In this paper, we proposed a photonic time-interleaved analog-to-digital conversion (TIADC) scheme by exploring optical clock distribution. A proof-of-concept experiment with a four-channel TIADC was implemented and demonstrated. Experimental results show that the optical clock distribution system can offer a four-channel clock with low jitter. These results are very encouraging toward the development of optical clock distribution based electrical systems in the practical applications.

Abstract:This paper propose a power supply modeling method to solve the long-distance power supply of marine seismic towed streamer acquisition equipment. Based on the characteristics of power supply structure, the paper establishes circuit model and recursive equation system, and constructs numerical model and forms analysis method combining with the actual parameters such as equipment power consumption, streamer length and power line resistance. Based on this method, a power supply system for marine seismic streamer acquisition equipment is successfully completed, which is also applied to the production of marine oil and gas seismic exploration. And through the actual data of streamer power supply unit, such as voltage and current, the results show that the actual data is consistent with the model analysis results, which verifies the effectiveness of this method.

Abstract:PUF(physically unclonable functions,PUF) is a new type of information security hardware,which has been widely used in the fields of Internet of things, consumer electronics and so on.SRAM based PUF is one of the most widely used types in industry.The SRAM PUF, based on Huahong 0.11μm CMOS process, solves the error rate problem caused by SRAM instability by introducing a BCH algorithm.Through the design of MCU test circuit and PUF chip test board, the key indexes such as in-chip hamming distance, inter-chip hamming distance and stability of PUF chip are tested and analyzed in detail.According to the test results, the hamming distance between the PUF films is 42.2% and the hamming distance within the films is 20.0%. When the error correction mechanism of the BCH algorithm is running, the hamming distance in the chip is reduced to 0, which solves the error rate problem in the process of PUF implementation. This study shows that the problem of instability can be solved by combining BCH algorithm with SRAM unit. The PUF chip combined with BCH algorithm and SRAM unit can meet the needs of recognition, electronic label and other applications.

Abstract:In the brain-computer interface technology, MI-EEG is difficult to extract features by hand and the classification accuracy is low due to its high degree of non-stationarity. This paper conducts corresponding research and exploration to solve this problem, and designs a shallow convolutional neural network (SCNN) based on short time Fourier transform (Short time Fourier transform, STFT). Through tuning experiments on the hyperparameters of the network model, the model was tested on the public data set 2b of the BCI competition in 2008 and got good classification results. The classification accuracy rate reached 80.23%, which was much higher than the accuracy of the CNN model without STFT processing of 61.04%. Under the same test indicators, it is superior to the traditional machine learning classification method (CSP+SVM) with a classification accuracy of 73.52%, and it also has more advantages than the same type of deep learning method (CNN-SAE, 77.60%).

Abstract:Aiming at the problem that there are a lot of Gauss white noise in PD signal of mine motor, a denoising method based on VMD and SVD is proposed. Firstly, the VMD algorithm is used to decompose the noisy signal; then the IMF component that meets the requirements is selected by the kurtosis criterion to reconstruct the signal; finally, the VMD reconstructed signal is denoised by the singular value algorithm to obtain a relatively pure PD signal. The experimental data show that the signal-to-noise ratio and mean square error of the PD signal obtained by the above methods are better, and the Gaussian white noise in the PD signal of mine motor can be removed more effectively, which achieves the expected effect and has certain engineering value.

Abstract:To solve the phenomenon that the existing Wi-Fi module will fail to connect when accessing to the wireless network, a joint distribution network scheme of "intelligent distribution network" + "soft AP" is proposed. Firstly, enter the intelligent distribution network, and use the listening mode of Wi-Fi module to capture the distribution network data for connection. If the intelligent distribution network fails, the Wi-Fi module will enter the soft AP distribution network mode, and the user will send the SSID and password of Wi-Fi for distribution network. After the success of distribution network, the system uses MQTT transmission protocol for background binding and registration. On this basis, aiming at the problem that the distribution network time of products is too long in the distribution network, the contract frequency is tested, and the contract frequency range with the shortest distribution network time is obtained. The test results show that the distribution network time of this method is within 4058ms, and the success rate of distribution network is improved, which can provide reference for the design work of IOT engineers.

Abstract:In the industrial heating furnace，due to the influence of high temperature environment, the general detector cannot measure the real temperature of the workpiece .Aiming at the temperature detection of industrial heating furnace, a real-time temperature measurement system of polysilicon infrared detector is designed. The relationship between the detector output and the ambient temperature, as well as the photoelectric conversion performance were tested by blackbody furnace, the results show that the output decreases sharply when the working environment temperature exceeds 30℃, with the increase of target temperature, the dynamic range decreases. The design adds an attenuator in front of the detector and a protective cover with sandwich structure, the results show that the system effectively expands the dynamic range of the polysilicon infrared detector and ensures that the ambient temperature of the detector is within 30 ℃, The measurement accuracy reaches 0.85 ℃, which meets the industrial requirements, and solves the application of polysilicon infrared detector in industrial heating furnace.

Abstract:Because the white blood cells, trichomonas and epithelial cells in the samples under the leucorrhea microscopic examination have the characteristics of fluidity, and they are on the two or more focal planes. In order to solve the problem that it is difficult to obtain the focal plane of interest, an automatic focusing definition judgment algorithm based on visual tracking is designed. The algorithm is divided into two focusing processes, coarse granularity and fine granularity. Coarse-grained focusing is mainly used to fix the focal plane near the true focal plane. Fine-grained focusing starts near the end position of the coarse-grained focusing. Fine-grained focusing is realized by moving the camera gradually, extracting the pathological basis object such as white blood cells and candida albicans, performing definition recognition judgment. Finally the clearest focal plane where the objects of the pathological basis are located is obtained. Through 1556 actual tests of the automatic focusing process, the algorithm can effectively improve the success rate of automatic focusing. By achieving automatic focusing, the cost of laboratory manpower is saved.

Abstract:The orthogonality between pulses of time-domain quadrature modulated signals of Prolate Spheroidal Wave Functions(PSWFs) is easily affected by channel characteristics.Among the channel compensation methods, the linear equalization method has limited compensation accuracy and the existing decision feedback equalization method has high training complexity for parallel signals. To solve these problems, a parallel adaptive decision feedback equalization method based on initialization parameter passing is proposed. The initialization parameter transfer method is used to realize the fast initialization of multiple parallel equalizers. The output weighted average mode conversion mechanism is used to quickly and accurately judge the switching time of decision-making mode.The simulation results show that the proposed method has fast convergence speed and low complexity, and the steady-state error is improved by 4 dB and 0.5 dB respectively compared with linear equalization method and traditional decision feedback equalization method.

Abstract:When particle swarm optimization (PSO) is adopted to optimize antenna array pattern, the optimizer’s convergence can be greatly improved by the efficient estimations of the optimum particle’s initial values, and some global optimal solutions better than usual can be obtained. By using the array’s Taylor weighting to estimate the optimum particle’s initial values, a new method is presented to improve PSO optimizer’s convergence. Comparisons are made between three methods of estimations, which are random weights, Taylor weights and analytic weights respectively, it can be said that Taylor weights and analytic weights are both effective to improve the PSO optimizer convergence relative to the random weights, and they both have similar performance. As the analytic weights only can be solved by performing large-scale matrix’s generalized inversion which is usually difficult and hardly afforded by common PC, while Taylor weights are relatively simple and easy to be calculated, the practicability of the latter is far better than the former.

Abstract:Measurement speed is a key fact that limited the practical application of RF device behavior model. In this paper, a novel rapid extraction method based on reverse wave injection is proposed. This method changes the amplitude and phase of the reverse incident wave at the load side of the device under test, so that massive training data can be acquired for the Bayesian algorithm, without the need for a complete load-pull test. Experiment is conducted over a GaN device at 2GHz frequency, the results prove that the method has greatly increase the extracting speed without compromising the accuracy. Compared with the traditional scheme, this method can greatly improve the model extracting efficiency and can be widely applied in practice.

Abstract:A multi-objective optimization method for energy management strategy (EMS) of the fuel cell hybrid electric vehicle (FCHEV) is proposed to improve the efficiency of the drive system and optimize the durability of fuel cell. The equivalent hydrogen consumption model of the hybrid power system is established according to the power flow and the efficiency characteristics of key components. In addition, the lifetime degradation of fuel cell based on the load variation is considered. The energy management system is achieved by presenting an intelligent power allocation method, that is, the control strategy based on fuzzy logic control (FLC). In further research, in order to ameliorate the energy management strategy, the parameters of the fuzzy controller are optimized with the assistance of genetic algorithm (GA). A multi-objective optimization problem which takes equivalent fuel consumption and fuel cell lifetime as optimization targets is proposed. The improved fast non-dominated sorting genetic algorithm (NSGA-II) is used to solve the multi-objective optimization problem, so as to optimize the control parameters. Finally, the optimization results of the above algorithm are tested, and the optimized strategy and other strategies are simulated by the advanced vehicle simulator (ADVISOR) under typical conditions. The results demonstrate that the optimization is effective, and the optimized control strategy has a considerable degree of superiority.

Abstract:With the continuous increase in the number of road vehicles, traffic accidents have become one of the main factors that endanger social public safety, and the prediction of road traffic accidents has also become a research hotspot.Taking into account the intricacies of accident influencing factors and the dynamic spatio-temporal variability and data sparseness of accidents, the fusion of multi-source data and the feature extraction according to time-varying and time-invariant data, especially the text description of the accident The feature extracts context information, and at the same time, the negative sampling method is used to balance the ratio of positive and negative samples.Finally, a regional traffic accident prediction network model trained by multi-feature component combination training is proposed. The model was validated on the data sets of three cities with different accident sparsity in the United States. The experimental results show that the prediction model is better than the basic model of comparison in various evaluation indicators, and each indicator is increased by about 2%3%. It can be seen that the model has improved the prediction performance to a certain extent. At the same time, the experimental results of different combinations of multi-feature components show that various factors have an impact on the occurrence of accidents.

Abstract:Aiming at the characteristics of high complexity and high accuracy of underground coal mine positioning. For improving the accuracy and efficiency of coal mine precise positioning and ensure the safety of personnel, an ultra wideband positioning algorithm based on improved TDOA is proposed. According to the different conditions of underground roadway, different training and filtering methods are adopted to ensure the accuracy and low complexity of underground positioning. The experimental results show that the average positioning accuracy of the combined algorithm can be maintained below 10cm in both LOS and NLOS, and compared with the basic TDOA algorithm, the average positioning accuracy is improved by 20cm and 35cm in LOS and NLOS, respectively; while in Los, the trained neural network is used for high-speed positioning to improve the positioning efficiency. Therefore, the combined algorithm can achieve the balance of positioning accuracy and efficiency in different environments, meet the requirements of high-precision positioning underground, and is suitable for underground personnel positioning.

Abstract:KITTI is a large data set fused with multiple sensors in automatic driving scene, its data acquisition platform includes two gray-scale cameras, two color cameras, a velodyne 64 line lidar, four optical lenses and a GPS navigation system. KITTI 3D Object Detection Evaluation can verify the accuracy and effectiveness of various 3D object detection algorithms. It is the most important data set in the field of autonomous driving. The focus of this article is the data reconstruction and data cleaning of the KITTI data set: first, use the RANSAC algorithm to remove the ground from each frame of lidar data in the KITTI data set, and use the DBSCAN algorithm to cluster the targets on the ground, and then according to the label The file uses the nearest neighbor search to assign tags to each target category to complete the data reconstruction. Based on this, the data is resampled to balance the categories to complete the data cleaning. For the reconstructed and cleaned KITTI data, the PointNet algorithm is used to complete the classification task, and the accuracy rate is as high as 95.13%. Finally, the overall framework of 3D target detection and evaluation on the KITTI data set is completed. The results show that the quality of the reconstructed and cleaned new data set is high, the classification algorithm is robust, and the 3D target detection process is clear and complete.

Abstract:Aiming at the problem of fuzzy load model of plug-in electric vehicle charging pile load increasing gradually in power grid, a data fuzzification process based on linear re projection algorithm is designed. Multi column deep convolution fuzzy neural network is used to analyze the data, and the inverse function of linear re projection algorithm is used to solve the data fuzzification. Finally, the input data in 24 h time sequence is used to push forward 24 h for high precision Degree prediction and estimation. The results show that the model effectively improves the management efficiency of electric vehicle charging grid, and is suitable for load management of plug-in vehicle charging pile.

Abstract:The new bone fixator is an external fixator that can be used to treat fractures, correct skeletal deformities and other conditions, and requires prescription software to calibrate and prescribe treatment. In this paper, a systematic calibration method is designed for a new external fixator based on binocular stereo vision. The method first calibrates the binocular image of the external fixator, finds the internal and external parameters of the camera and corrects the image, then uses the principle of binocular stereo vision to solve for the coordinates of the key points in the processed image and uses the key point coordinates to calculate the length of the fixator linkage. The method was validated to be able to control the error within 05.mm. The data calculated by this method can be used directly by the prescription software when designing the treatment plan, which is important for optimising the process of using the new external bone fixation frame.

Abstract:3D medical image registration algorithm is widely used in scientific research, follow-up and other medical fields, so it is of great significance to improve its registration accuracy. Aiming at the problem of medical image registration, a V-shape network based on V-Net(VV-Net) is proposed. The registration model can be trained end-to-end by stacking V-Net. Specifically, the moving image is distorted by two V-Nets in turn, and the additional V-Net is used to provide supplementary information for the first two V-Nets to form a V-shaped network, so that the moving image can be better aligned with the fixed image. At the same time, the depth supervision auxiliary branch is added to the proposed model to prevent over fitting. The accuracy of registration is improved by using the progressive registration and information supplement. The performance of the model is evaluated by ADNI、ABIDE、ADHD200 and OASIS data sets. Compared with affine transformation, symmetric normalization (SyN) and VoxelMorph, the proposed registration method achieves 24.7%, 13.2% and 1.3% accuracy improvements, respectively. The experimental results show that VV-Net has achieved good results in the field of medical image registration.

Abstract:Aiming at the problem of existing monocular vision measuring technique in synchronously measurement for structural 3D vibration and rotation parameters measurement, a series of novel vision displacement-sensitive fringe pattern (DSFP) based dynamic measurement methods were proposed. Multi-dimensional and multi-parameter synchronous measurement for non-rotating structural 3D displacement, rotating shaft 3D displacement, rotation angle and rotation speed could be realized by using the novel DSFPs as vibration displacement or rotation angular displacement sensors and using a high-speed camera as a detector. The evolution and measurement principle of different DSFPs were well descried. The feasibility and accuracy of the vision DSFP-based measurement methods for the measurement of non-rotating structural 3D displacement, rotating shaft 3D displacement and rotation parameters were verified by experimental results, which provides a series of measurement solution for structural condition monitoring and faults diagnosis.

Abstract:In order to solve the problems of low coverage balance performance, low quality of life of nodes and serious fluctuation of network transmission performance in the process of wireless sensor network deployment, a WSN balanced coverage algorithm based on weight threshold liberalization mechanism is proposed. Firstly, the infection mechanism is introduced and the network nodes are regarded as infection particles. A weight optimization method based on redundant coverage strategy is designed. The redundant coverage coefficient is calculated according to the weight coefficient of each particle, and the nodes with increased redundant coverage coefficient are shifted to improve the evaluation ability of the network on the degree of redundant coverage. Then, combined with the initial coordinates, maximum coverage radius, redundant coverage weight and other coefficients of the node to build a decision model, a node mobile coverage method based on threshold decision mechanism is designed. The node variance is verified by the time-lapse coordinates, and the decision threshold is further constructed by the node variance. The nodes whose decision threshold is not higher than 1 will not be moved, which reduces the loss due to movement Network coverage is reduced due to network failure. Simulation results show that: compared with WSN coverage optimization algorithm based on global local search mechanism and WSN robot coverage algorithm based on improved filtering mechanism, The proposed algorithm has higher node survival rate and network transmission bandwidth, which reach 90% and 3780mbps respectively in Gaussian channel.

Abstract:In the traditional multi-point distributed temperature measurement system, there are problems such as complex wiring, increased cost and real-time data processing requirements. Therefore, this paper designs a system that uses Bluetooth Mesh networking technology to achieve multi-point wireless temperature measurement. The system uses ARM as the core development board, PT100 as the temperature sensor head, uses 16-bit high-precision analog-to-digital converter ADS1110 to collect the temperature signal output by the PT100 temperature sensor, and realizes the wireless transmission of the temperature signal through the Bluetooth spiral antenna. Use the LabVIEW interface to display the temperature change curve in real time, while temperature alarm and temperature data can be saved. A method combining the least squares method and digital filtering is applied to the calibration of temperature data to eliminate the influence of system errors and interference signals and obtain high-precision temperature values. The experimental test results show that the temperature measurement system has the advantages of low cost, low power consumption, high precision, and convenient operation. At the same time, the temperature measurement accuracy can reach ±0.1℃, which can be widely used in the real-time acquisition and processing of temperature signals. It has certain practical value in it.

Abstract:This paper develops a standard time reproduction terminal based on GNSS satellite common-view. The technology used by the terminal has improved the traditional satellite common-view, so that the common-view period can be flexibly configured according to the type of clock source used by the users, and is not limited 16 minutes in the traditional common view.In addition, a set of remote reproduction system is built using this technology. The system is composed of a reference station and a user station. A standard time reproduction terminal is placed in both places. The terminal is mainly composed of a satellite card, a time interval counter, and a clock source. Features such as small size, low cost and flexible composition. By using the common view comparison result of the base station and the user station to adjust the clock source output signal of the user station, the time and frequency signal synchronized with the base station can be reproduced locally in the user in real time, and the absolute deviation of synchronization can be kept within 10ns (95%). For users, there is no need to deeply understand the implementation principle, only a reproduction terminal is needed to obtain a time-frequency signal synchronized with the time of the reference station with high precision, thus meeting the needs of users in all walks of life for time-frequency signals . According to the test experiment, it is concluded that the time-frequency signal of the dual-frequency version of the reproduction terminal has a maximum deviation of 7.19ns from the national standard time UTC (NTSC) frequency signal, the standard deviation is 1.49ns, and the frequency accuracy is 4.25e-14; The maximum deviation between the time-frequency signal recovered in the remote location of the single-frequency version of the terminal and the national standard time UTC (NTSC) frequency signal is 10.9ns, the standard deviation is 4.23ns, and the frequency accuracy is 3.79e-14.

Abstract:In order to solve the problem of single measurement type of reading instrument for dam safety monitoring, a new type of reading instrument for safety monitoring is designed, which can efficiently realize the multiplexing measurement of the two kinds of sensor signals, vibrating wire and differential resistance. It uses STM32F407VET6 as microprocessor, through vibrating wire signal acquisition, differential resistance signal acquisition, channel multiplexing, Bluetooth communication, power control, human-computer interaction and other unit modules, to realize sensor signal measurement, mode switching , measured value displaying and so on .The experimental data and results show that the designed readout can realize the multiplexing measurement of vibration wire sensor and differential resistance sensor, and its absolute error parameters meet the current standard requirements.

Abstract:In order to accurately measure the dielectric loss of electronic materials at a specific microwave frequency, a dielectric loss measurement method based on the defect model of microwave photonic crystal is proposed. Considering the microwave energy localization of the defective microwave photonic crystal, the dielectric under test is designed to be a defect layer within the microwave photonic crystal. The monotonic relationship between the transmission defect peak value and dielectric loss factor of the measured material is calculated by using transfer matrix method. When the dielectric loss factor varies from 0 to 1.0, the defect transmission peak value reduces from 0 to 0.21. As a result, when the defect transmission peak value is tested, and the dielectric loss factor can be obtained by using the calculated relationship between the transmission defect peak value and dielectric loss factor.

Abstract:Aiming at the problems of low efficiency, complicated data processing and human error interfer-ence in the manual verification of Sound Level Meter, an automatic verification system for the electrical performance of Sound Level Meter is proposed. The system taking multi serial port industrial computer as the control core, use LabVIEW actor framework and object-oriented technology, the HAL and MAL areconstructed to complete the packaging of verification process,verification index, standard instrument and equipment, and the sound level meter to be tested.Each functional module is built through the "class" torealize the automatic control of instrument,data acquisition, data analysis and processing, and original data recording in the verification of sound level meter Record and other automatic verification work. After many tests, the system works stably, and can automatically complete all the verification items of the So-und Level Meter, which is consistent with the traditional manual verification results. The verification timeis reduced to half of the original, and the frequency range of the sound level meter is 1 Hz ~ 200 kHz. Keyword：metrology;automatic verification；Sound Level Meters;Labview; Actor Framework

Abstract:Underwater robot (RemoteOperatedVehicle, hereinafter referred to as the ROV) underwater hovering is a very important shipment.In order to achieve the underwater hovering control, we design a hovering control method based on multi-sensor information fusion. The ROV respectively calculated by accelerometer and gyroscope and depth for the body of the shaft, attitude Angle and depth of acceleration signal of the gyroscope data by quaternions, get the body attitude Angle in real time, through the weighted sliding filter of gyroscope and accelerometer data beforehand. After using kalman filter for accelerometer and gyroscope data fusion, get real-time acceleration. Finally, BP neural network to forecast the depth and the attitude Angle, and the prediction of depth and the attitude Angle into the PID controller as the control strategy. The fusion method can effectively complete the hovering control, and the experimental results show that this method can achieve higher stability and real-time performance, has a broad prospect in the hovering control.

Abstract:In the paper，An indoor pollution gas sensing and evaluation system based on Stacking integrated learning is designed. Users can use mobile phone APP scan the QR code of the detection terminal to view the real-time data and evaluation results of the current environmental pollution gas. By using Stacking, the system integrates evaluation algorithm with classification algorithm to solve the problem of strong subjectivity of evaluation algorithm. Firstly, fuzzy mathematics comprehensive evaluation method, decision tree and KNN are selected as the basic models for training, and then the output results of the three basic models are taken as eigenvalues, and the logical regression is used as the meta-model to integrate the three basic models heterogeneously. This method can improve the accuracy and objectivity of the evaluation results while ensuring the real-time performance of the system.