Abstract:The movement of slide of large table hot die forging hydraulic press is interfered by many factors. The working plane of slide is prone to tilt, which seriously affects the service life of forging die and product quality. It is necessary to conduct real-time tracking measurement and rectification. In view of the shortcomings of the exist slide balance detection methods for large-scale hot die forging press, based on the tilt angle of the moving slider, a dual state detection method of the whole tilt and the deformation of the lower surface is proposed, and designing the sensor node layout that is suitable for this method. Based on ZigBee wireless sensor network, a balance state remote monitoring system of moving slide has been presented, which can detect the balance state of the moving slide in real time. Through the operation experiment of the slide of the large table die forging hydraulic press, the comparison between the remote real-time detection data and the actual working data of the slide shows that the detection accuracy reaches 0. 01 °, The results show that the detection method of the balance state of the slide and the remote monitoring system have certain reference significance for the normal operation of the hot forging hydraulic press.

Abstract:Aiming at the existing neural network fault location algorithms for ground faults on VSC-HVDC lines, there are too many training samples, long training time, and no effective verification of robustness is proposed. A method based on ST and PSO optimizes the line fault location algorithm of GRNN. From the perspective of the fault traveling wave energy spectrum, the ST is used to extract the fault transient voltage signal energy spectrum, and the energy representing each frequency interval is summed to achieve accurate extraction of the energy characteristic samples; and then normalized the subsequent energy samples and input to the neural network for training, and the PSO algorithm is used to optimize the smoothing factor of the GRNN to improve the network convergence speed and training accuracy. Finally, the electromagnetic transient simulation proves that the method has high positioning accuracy and is not easily affected by the transition resistance. In the case of input samples with measurement errors and external noise interference, the maximum error is still less than 1. 5%, which has certain engineering application value.

Abstract:As a key part of system energy supply, the end of life of lithium-ion battery often leads to the degradation or failure of the electrical equipment, or even the collapse of the whole system. Therefore, it is increasingly important to study the remaining useful life (RUL) of the battery and predict the failure time in advance. Aiming at the problems of long training time, difficult parameter determination and unstable output results during the life prediction process of lithium-ion batteries, this paper puts forward the Relevance Vector Machine (RVM) which is more suitable for on-line detection and has better generalization ability, sparser parameter and shorter test time, and optimizes relevance vector machine (RVM) through quantum particle swarm optimization (QPSO) to ensure the stability of predicted output results. The results show that the prediction accuracy of the improved relevance vector machine (QPSO-RVM) is up to 99%, the mean absolute error of battery life prediction is about 2% and the root mean square error is about 3%, which verifies the improved algorithm feasibility and superiority.

Abstract:In order to solve the problems of traditional methods, which do not consider the influence of multi parameter indexes on power quality and low accuracy of power quality measurement results, a power quality measurement and evaluation model based on RFID technology is proposed. The power quality EPC code is collected by using RFID technology to obtain the power quality measurement and evaluation data, and the total harmonic distortion rate of current, three-phase imbalance and frequency deviation are taken as the measurement and evaluation indexes to calculate the weight and obtain the comprehensive weight value of each index; the radar chart analysis method is introduced to compare the multi variables of power quality, and the comprehensive evaluation function is defined to reflect the power quality to achieve comprehensive power quality assessment. The experimental results, which take the evaluation stability and accuracy as the experimental indicators, show that the model can achieve high-precision evaluation of power quality. The reliability of practical application is strong.

Abstract:Effective prediction of wireless communication link quality is a necessity to choose the reliable routing of multi-hop Internet of things (IoT) communication. The main challenge for its inaccurate prediction is caused by the random characteristic of the signal-tonoise ratio time series. To address this problem, based on the analysis of the random characteristics of wireless communication links, a method of predicting the confidence interval of communication quality is proposed in this paper. Firstly, the signal-to-noise ratio time series of wireless link quality is decomposed into stationary sequence and noise sequence by wavelet decomposition method. The noise standard deviation sequence is obtained by the noise sequence. Then, the prediction model of stationary sequence and noise standard deviation sequence is proposed by using LSTM neural network. The confidence interval of communication link reliability is calculated by using the prediction results. Finally, by comparing the lower bound of confidence interval with the reliability standard, it can prejudge whether the reliability of current wireless link meets the requirements of power grid. Through the comparative study, the proposed method can either satisfy the requirements of the application of IoT of distribution grid or provides more accurate result in comparing with the state-of-the-art methods.

Abstract:A communication architecture based on backscattering radio and software definition is proposed for WSNs which require low bit-rate and ultra-low cost sensors. It consists of a central Hub for RF carrier generation, homodyne detection and complex information processing, and sensors with carrier modulation function. In the former, a software-defined transceiver is constructed to transmit the carrier and receive the reflection from various sensors, and extract and process their transmitted messages. In the latter, the transmitter for each sensor is simplified as a transistor connected to an antenna and each sensor is designed around a low-power micro-controller driving a low-power RF switch, and the information from each sensor is modulated onto its unique subcarrier. Thus, a complete backscattering radio link communication system is established to achieve communication between each sensor and the center Hub. The simulation results based on the theory and the experimental results from the prototype show that the proposed WSN communication system not only achieves a wide range of reliable communication and data transmission and processing, but also has low complexity and ultra-low cost.

Abstract:In order to quantitatively characterize the degree of deformation mismatch of adjacent layers of composite flexible electronics, an experimental system based on digital image correlation (DIC) and a method of evaluating deformation mismatch are designed. First, a composite flexible electronic sample was prepared, and the experimental system was verified through translation experiments. Second, the sample was uniaxially stretched at different temperatures and the Von Mises strain was calculated. Finally, the degree of deformation mismatch was Characterized by defining the patterned strain fluctuation index P. The experimental results showed that the translation error of the experimental system was within 1%, which meets the requirements of deformation measurement. The sample was stretched by 2 mm at 50 ℃ , and the maximum and minimum strains of Von Mises differed by about 65%. The sample showed deformation mismatch with uneven strain value but uniform strain distribution. Under the same tensile conditions, the degree of deformation mismatch changed little from 25 ℃ to 75 ℃ , and it changed significantly at 100℃ , the increment was about 60%. This method effectively characterizes the degree of flexible electronic deformation mismatch.

Abstract:Vehicle-to-vehicle (V2V) communication system is an important part of the future intelligent transportation system, and its non-stationary characteristics have been verified in channel measurement. A non-stationary double-ring geometry-based stochastic model is proposed for V2V multiple-input multiple-output communication systems. In order to describe the non-stationarity of the channel, the time-varying characteristics of the angle of arrivals (AOAs) and the angle of departures (AODs) are introduced into the model, the time-varying statistical characteristics such as the spatial cross-correlation function (CCF) and the time autocorrelation function (ACF) of the model are derived. In addition, based on the modified equal area method ( MMEA), the corresponding simulation model is developed. The effects of scattering environment, moving state of mobile terminal and antenna deflection on channel statistical characteristics and non-stationarity are studied. Simulation results show that the proposed model can effectively simulate the nonstationarity of the V2V channel and the statistical characteristics of the channel under stationary state are consistent with those of the stationary double-ring model, which verifies the correctness of this model.

Abstract:Aiming at the problem that the capacity regeneration phenomenon affects the state of health ( SOH) prediction accuracy of lithium-ion batteries, an energy-weighted Gaussian process regression (EWGPR) of empirical mode decomposition (EMD) method is proposed. This method regards the capacity recovery phenomenon as the energy projection of the capacity decay process of lithium-ion battery. The energy distribution is obtained by EMD decomposition and the sample weights are calculated according to the energy distributions. Then the SOH prediction model of lithium-ion battery based on EWGPR is established. The experimental simulation results on the NASA lithium-ion battery datasets show that the EWGPR algorithm has higher accuracy and adaptability than the basic GPR algorithm, and the root mean square error (RMES) for single-step and multi-step predictions are decreased by more than 3% and 10%, respectively.

Abstract:In recent years, generative adversarial nets ( GANs) have developed rapidly and have become one of the main research directions in the current machine learning field. GAN is derived from the idea of zero-sum game. Its generator and discriminator are opposed to learning. The purpose is to obtain the data distribution of a given sample and generate new sample data. A large number of investigations have been made on GAN models in image generation, abnormal sample detection and location, text generation pictures and picture super-resolution. The substantial progress made in the application of these GANs has been systematically explained. The background and research significance, theoretical model and improved structure of GAN, and its main application fields are summarized. The shortcomings of GAN and its future development direction were summarized.

Abstract:For the practical application requirements of location-based services, a sparse fingerprint localization method based on spatial position constraint is proposed, after fully analyzing the limitations of the existing indoor location technologies. The positioning information from inertial navigation system (INS) and wireless local area network (WLAN) are deeply integrated on the data level, to coordinate the positioning task. Based on the received signal strength (RSS) data provided by WLAN, the spatial-location-fingerprint database is constructed, together with the sparse fingerprint representation and location model. In view of the RSS variability due to environmental interferences, the displacement state can be preliminarily estimated by INS, which will be as a constraint condition to construct the sparse fingerprint location model based on spatial position constraint. The simulation experimental results show that the positioning accuracy of this method is improved by 58% and 33% respectively, compared with the INS and sparse fingerprint methods. It is demonstrated that the proposed model can appropriately compensate the accumulative error of INS, and the motion prediction by INS also can restrict the jumping and distortion effects of RSS signals to a certain extent.

Abstract:Aiming at the problems of damage caused by short-circuit faults and low fault identification rate of transmission lines, a fault identification method combining VMD-PE and siamese neural networks ( SNN ) is proposed. For determining the number of decomposition layers K, use the instantaneous frequency mean to optimize VMD parameters, decompose the three-phase voltage at fault by VMD, calculate the permutation entropy of each component after decomposition, and use them as the fault features; input the fault features into the trained SNN for similarity measurement, compare the similarity between the two input samples to determine the type of short-circuit fault on the transmission line. The feasibility of the method is verified by simulation experiments, and compared with other classification methods, the accuracy and superiority of the method are proved.

Abstract:Aiming at the problem of low accuracy of lane detection in complex environment, a lane detection algorithm based on directional distance transform coupled with multi-particle filter was proposed. Firstly, the four-point perspective mapping method was used to transform the input image into an aerial view, which makes the lane boundary parallel and convenient for lane detection. Oriented distance transform (ODT) was introduced to mark the edge pixels of aerial view to the nearest points in horizontal and vertical directions to find the initial boundary points. Secondly, the lane model was constructed by using the lane center, the angle from the center to the left and the right boundary and the tangent angle of the left and the right lane boundary. Two independent 4D particle spaces were applied to the left and the right lane boundary. Subsequently, a multi-particle filter is introduced into the lane model to detect and track a pair of lane boundary points using particles propagating independently on both sides of the lane, and the boundary points are adjusted by local linear regression. In order to optimize the performance of multi-particle filter, dynamic dependencies were created according to the particle state vector. Finally, the weight of particles is determined by iteration, and the lane line was detected by multi-particle filter. Experiments show that, compared with the current popular lane detection algorithms, the proposed algorithm has higher detection accuracy and robustness in a variety of complex interference environments.

Abstract:Aiming at the problem that Lamb wave is difficult to quantify the damage of aluminum plate, a short-space two-dimensional Fourier transform method is used to study it. Firstly, the Lamb wave is excited in aluminum plate by air coupled ultrasound and its timespace wave field signals is obtained by linear scanning. Then, the short space two-dimensional Fourier transform is used to obtain the Lamb wave space-wavenumber curve of the scanning area, and the location, length and depth of the damage are obtained. Finally, the depth of damage is deduced from the phase velocity frequency-thickness product curve. The results show that the proposed method can simultaneously evaluate the plate thickness of the scanning area and the location, length and depth of the damage. Among them, the maximum quantitative error of plate thickness is 5. 50%, and the maximum quantitative error of damage length and depth is 6. 00% and 6. 67%.

Abstract:Aiming at the problem of high-order amplitude phase shift keying ( APSK) demapping complexity and difficult hardware implementation, a low-complexity APSK demapping scheme and circuit implementation architecture are proposed. Specifically, the constellation map is divided into regions based on the analysis of symmetry. Then, based on the Max-Log-MAP algorithm, the bit soft information of the received symbols falling into each region is calculated and simplified, thereby obtaining a formula with a low calculation amount for calculating the soft information. Furthermore, using the characteristics of the simplified soft information calculation formula for each bit, the soft information calculation circuit architecture is designed and its performance is tested on the field programmable gate array (FPGA) hardware platform. Test results show that the APSK demapping circuit using the proposed simplified method can achieve a bit error rate (BER) of 10 -5 when the signal-to-noise ratio (SNR) is 14 dB, which is close to the performance of the traditional demapping algorithm and has lower hardware resource consumption.

Abstract:Aiming at the issue of poor robustness of multiple-input multiple-output (MIMO) radar detection caused by the initial clutter estimation error, a joint robust optimization approach of transmitted waveform and received weight is proposed here to improve MIMO radar detection robustness. With the constraints of clutter error convex set, the transmitted waveform constant envelop characteristic and the similarity, the min-max joint optimization problem can be firstly constructed to improve the worst-case detection performance of MIMO radar on the basis of the criterion of maximizing the output signal to interference noise ratio (SINR); After that, in order to solve the resultant NP-hard problem, this issue is decomposed into the internal and external sub-problems, and these two sub-problems can be solved alternately. In comparison with the non-robust and existing robust algorithms as well as unrelated signals, numerical simulation verifies the efficacy of the developed approach.

Abstract:In recent years, significant progress has been made in two-person interaction recognition based on RGB video, but there are still many problems in RGB video data that seriously affect the recognition rate of two-person interaction. With the rapid development of depth sensors (such as the Microsoft Kinect), it is possible to directly obtain a data point that can track human movement, making up for the lack of RGB video data. Therefore, a two-person interaction behavior recognition method based on node data is proposed. First, HOJ3D features and joint distance features were calculated from the data of the node, and then were graphically sent into different convolutional neural networks. Then, the two features were extracted and splicedtogether. Then, Softmax classifier was used for classification and recognition. The test results of the method on the SBU Kinect action dataset show that the recognition accuracy of the method has been improved to a certain extent, reaching 94. 4%. The method is simple to implement, has the ability of real-time processing, and has a good application prospect.

Abstract:Aim at the difference in trajectory tracking accuracy during the movement of smart cars. An intelligent vehicle trajectory tracking algorithm based on inversion control algorithm is proposed. First, establish the kinematics model of the smart car, error model and dynamic model. Then, according to the inversion algorithm, design a reasonable amount of virtual control of the branch, combined with Lyapunov stability analysis, the intelligent vehicle trajectory tracking control law is designed. Finally, the simulation experiment of intelligent vehicle trajectory tracking control is carried out on Simulink. The experimental results show that the designed smart car motion trajectory tracking algorithm has better real-time performance and better tracking accuracy in the above learning algorithm or Lyapunov direct method. And meet the needs of the stability of smart cars under different speed environments.

Abstract:There are two ways to solve the problem of midpoint potential fluctuation in the academic world: One is small vector adjustment based on SVPWM control; the second is zero-sequence voltage injection based on SPWM control. The midpoint potential balance control method used in this paper was an improved zero-sequence voltage injection method with correction control. Compared with the traditional zero-sequence voltage injection method, it took into account the advantages of traditional saddle-shaped modulated waves, and improved the over-modulation problem, and studied the practical zero-sequence voltage calculation method, which finally effectively suppresses the fluctuation of the midpoint potential, and the practical calculation method of checking and correcting zero sequence voltage is studied, the fluctuation of medium voltage point is successfully limited to 5 V, and the THD of grid side current is correspondingly reduced by 1%. Finally, the correctness and feasibility of the control method are verified by MATLAB/ Simulink software simulation and experiment.

Abstract:Aiming at the problem of the time-consuming and labor-intensive manual inspection, a MV ( machine vision)-based cable conductor quality inspection method is proposed, which can inspect the cable conductor quality conveniently and efficiently. After analyzing the characteristics of the knife mark in the cross-section image of the cable, a method based on Gabor filter is proposed to eliminate this texture. According to the distribution rules of cable conductors, a hierarchical analysis algorithm based on clustering is proposed. The experimental results demonstrate that the proposed method can effectively improve the effect of conductor contour detection, and implemented the layered inspection of conductor number, which can help inspectors quickly inspect the number of conductors and find the defects.

Abstract:In order to avoid the color distortion caused by the brightness enhancement of the image and the problem of over-enhancement in the local area, an image enhancement algorithm based on the image classification coupled adaptive Gamma correction (AGC) was designed to improve the image details and visual effects. Firstly, the input image was converted into HSV space and the color and the brightness are separated, so that the original color of the pixel was not changed when the brightness channel was enhanced, and the color distortion is effectively reduced. Secondly, considering the properties of different images, the images are classified into high and low contrast by using statistical information, and each contrast was divided into light and dark. Then, based on the traditional Gamma correction method, an AGC was formed by dynamically setting parameters for different types of images, thus, different enhancement functions are constructed for different types of images to complete the enhancement of different types of images. The experimental data show that compared with the current popular enhancement algorithms, the proposed algorithm has higher enhancement effect, which presents more natural brightness and contrast, as well as maintains more color information.

Abstract:Among many indoor positioning methods, WiFi based indoor positioning has great potential use in location based Service (LBS) applications. A new WiFi based positioning method is proposed in this paper, which adopts the prior information of user walking patterns. The method can also effectively adopt the traditional hierarchical positioning structure. In the pre-coarse positioning phase, the method can adopt the position estimation from the previous time to define the potential areas. In the signal fingerprint test phase, the potential areas can be tested to know whether it is right or wrong. If wrong, the second coarse positioning phase is carried out to re-define the potential areas. Then in the fine positioning phase, the maximum posterior (MAP) based method is adopted to estimate the user position. The proposed method is tested using an open access dataset, the results have shown that the proposed method can effectively adopt prior information and can increase positioning accuracy by 3%, 5% and 7% over the traditional methods ( no prior&MAP, prior&kNN and no prior&kNN) respectively.

Abstract:A micro-doppler period estimation method for ship target is proposed. Under the influence of sea conditions, ship targets are in the micro-motion state, which induces micro-doppler modulation on the radar echo. The micro-doppler parameters of the ship target are closely related to ship structure, target dynamics and so on, which is an important basis for ship target recognition. In this paper, based on the radar echo model of ship target, the modulation characteristics are analyzed, and a fast estimation method of micro-Doppler period is proposed. Firstly, the main body translation of ship target is successfully compensated based on the minimum entropy method; then the Doppler range is located according to the entropy difference between the micro-Doppler area and the noise area; finally, the timefrequency correlation coefficient is calculated to provide the micro-Doppler period estimation. The location of micro-doppler range reduces the computation burden of the time-frequency correlation coefficient, which makes it fast compared to the original algorithm. Under the typical scene and radar parameters, the efficiency of the method is improved by 2. 5 times.

Abstract:In order to solve the problem that invalid data affects the accuracy of EEG classification, a method of data screening is proposed. Based on brain computer interface (BCI) system, this paper presents an approach that using BP neural network to classify the EEG data generated by visual stimulation. The statistical characteristics of EEG signals corresponding to left and right motor imagery tasks are input to the BP neural network. First, the invalid data are eliminated by using the energy characteristics of β rhythm signal, and then classified by combining the mean value, standard deviation, energy spectrum, power spectrum, autocorrelation function and other features of μ rhythm signal. The using of β rhythm signal makes the characteristics more accurate and improves the accuracy of signal classification from 78. 25% to 84. 11%.

Abstract:When the non-invasive stimulation such as transcranial electric stimulation locks phases with the intrinsic neural electrical activity in the brain, the neural oscillatory activity can be regulated in a more effective manner. Due to the complex time-variation of EEG signal, the existing methods cannot meet the accuracy of phase estimation and real-time performance of the system at the same time. In this paper, a real-time phase estimation method for phase-locked stimulus system was proposed. In this method, the EEG signal was modeled by autoregressive (AR), then the AR model was used to predict the EEG signal and identify the phase feature points, and the phase to be stimulated was calculated by the phase feature points. The method was used to analyze the closed eye resting EEG of 20 subjects (aged 20~ 36, male 12, female 8) and it was found that the performance of the method is related to the updating time of the model coefficient, the prediction step and the narrow-band power of the EEG. It had better performance for the EEG with higher narrowband power. Under the optimal model parameters (the updating time of the model coefficient was 5 s and the predicted step length was 30), the average phase locking value (PLV) of the 20 subjects was 0. 968, and the average phase error was 13. 33. Compared with the average period method, this method has higher PLV value and lower phase error, which can be used in the development of closed-loop phase-locked electric stimulator.

Abstract:Aiming at the high-order reconstruction and evaluation of air combat maneuvers, this paper takes the flight state data of aircraft as the research object, and constructs the maneuver decision reconstruction index model by introducing the concept of jerkiness. In order to reduce the impact of subjective evaluation method on the evaluation results, this paper introduces the degree of jerk theory in physics, USES principal component analysis (PCA) to determine the weight of each index, and obtains the comprehensive evaluation value. Then the synthetic reconstruction function of maneuver decision point is constructed and the maneuver decision point is extracted. Based on the situation function and maneuver decision points, the objective data recorded in the air combat training are compared and analyzed.

Abstract:In order to promote the work efficiency, energy consumption and the smoothness of the manipulator, a mathematical model of quantic non uniform rational B-spline curve (NURBS) was set up to find trajectory with high-order of which endpoint parameters was specified. A hybrid particle swarm optimization algorithm ( HPSO) was used in the MATLAB simulation to get the perfect Pareto solutions under three normalized weighted objectives for six-degree freedom manipulator’ s trajectory. Through the simulation, it shows that the quantic NURBS curve can be constructed to fitting high order trajectory and the HPSO algorithm can provide a good means to get the Pareto solutions for the trajectory described above.