Abstract:Abstract:The gravitational and environmental stray magnetic fields will introduce measurement noises and system errors during atom interferometric measurement process. Usually, the magnetic shield material sheets are rolled and welded to become the cylindrical segments, which are spliced to form the cylindrical magnetic shield device with enough length. To clarify the influence of this fabrication method on the magnetic shielding performance, the variations of residual magnetic field distribution in shielding area caused by the number of shield layers, the axial and transverse weld seams and adding caps on both ends of the shield device are analyzed in detail in this paper. The results show that the existence of the axial weld seam with the width of 4 mm has little influence on magnetic shielding, however the presence of the transverse weld seam with the width of 4 mm on the shield with lower relative magnetic permeability will introduce obvious magnetic flus leakage, adding caps on both ends of the shield device may reduce the residual magnetic field intensity, but it can hardly enlarge the region of uniform magnetic field. Based on the analysis results, a duallayered cylindrical magnetic shield device used for the interference area of the upwardprojectile cold atominterferometry gravimeter was fabricated, and the actual measurement results reveal that a uniform magnetic field region with length of 700 mm and unevenness of 4 nT is obtained after demagnetization, which satisfies the requirements of atom interferometric measurement, and has good consistency with the simulation results. .txt

Abstract:Abstract:Infrared thermal imaging temperature measurement system has developed rapidly due to its characteristics of long distance, noncontact, multitarget and high accuracy. In current epidemic period of 2019novel coronavirus (COVID19), infrared thermal imaging temperature measurement technology highlights its importance in order to quickly and efficiently screen out hightemperature population, prevent and control epidemic situation in time. During the measurement process, the temperature of the measured object, ambient temperature, atmospheric temperature, lens temperature, detector temperature and other factors have certain influence on the temperature measurement accuracy, especially the highaccuracy calibration of the temperature measurement curve has the most critical influence on the temperature measurement accuracy of the infrared thermal imager. In this paper, the principle of infrared temperature measurement is studied, a matching model between the output gray value of the infrared temperature measurement system and the measured temperature of the measured object is established, and a method for accurately calibrating temperature measurement curves is found. The V-T4r curves applied to highaccuracy temperature measurement and the V-TFPA,V-Tlens and V-Tshutter curves applied to temperature drift compensation are proposed. Through experiment verification, the ideal effects were achieved, and the temperature measurement accuracy is significantly improved, which can reach within 015 K. .txt

Abstract:Abstract:The reconstruction measurement technology based on monocular structured light is widely used in modern measurement field, especially in the measurement of small and mediumsized lowreflective parts with certain curved surface characteristics, due to its high accuracy and low cost. The measurement error is an important factor affecting measurement accuracy. Firstly, according to the constructed monocular structured light reconstruction measurement system, the main sources of errors affecting measurement accuracy are analyzed, and the quality of the collected structured light pictures is ensured through reasonable setting and adjustment of the parameters of the hardware equipment. Then, in order to improve the sinusoidal property of the projected structured light, a simple method to reduce the nonlinear gamma response of a common projector is proposed. Finally, a Zhang Zhengyou optimization calibration algorithm based on circular calibration plate is proposed to improve the calibration accuracy of the system. The experiment results show that the accuracy of reconstructing the measured point cloud data on the surface of small and mediumsized curved parts is between 03 and 06 mm, and the reconstruction measurement time of a single projection on the surface of the measured object is within 2 minutes, which verifies that the compensation method is effective and feasible. .txt

Abstract:Abstract:Aiming at the measurement requirements of high radiation and attached space microthrusters, such as alpha decay microthrusters, a new vacuum dynamic microforce measuring platform based on high temperature superconducting maglev device is proposed in this paper. An experiment condition of low temperature, low pressure and low resistance is established by liquid nitrogen, molecular pump and high temperature superconductors, which avoids the existence of contact resistance. In the experiment, the hard magnetic rotor will suspend on the superconducting guide rail for a single degree of freedom plane rotation. A photoelectric speed measurement system is used to perform long term monitoring on the motion state of hard magnetic rotor remotely, the micro force measurement is achieved and the possible radioactive hazard is reduced. The paper introduces the key subsystems of the microforce measuring platform, analyzes the motion characteristics and force measuring mechanism of the microforce measuring platform. In addition, the experiment flow chart and data processing method are given. Calibration test was conducted, and after calibration the platform can realize the measurement of the thrust less than 1 μN. The resistance of the platform is as low as 511±35 nN, and the bottom boundary of micro force measurement is 66 nN. .txt

Abstract:Abstract:With the construction of global navigation satellite system (GNSS), the technology of multisystem fusion time comparison has become the developing trend in the field of timekeeping. In this study, the data of GNSS receivers are utilized, which are from the time keeping laboratories of National Time Service Center of Chinese Academy of Sciences, Institute of Photonics and Electronics Academy of Sciences of the Czech Republic, and SP Technical Research Institute of Sweden. Based on the multisystem measurements data and the precision products clock, orbit downloaded from the international GNSS service center, the multisystem GNSS data fusion technology in PPP time comparison is studied. Experimental results show that the multisystem fusion technology can increase the number of available satellites. Compared with the single system, the number of observed satellites is doubled. The effects of multipath error and observation noise are reduced, which are caused by observation altitude angel. The influence of receiver clock error is suppressed, which can be used to improve the stability and reliability of time comparison. With the multisystem GNSS fusion technology in the long baseline time comparison, the stability results are better than of the any single system. Compared with BDS, GLONASS and Galileo, the stability is increased by at least 5%. .txt

Abstract:Abstract:As the core component for robot posture control, smallsize joints have a direct influence on the robot performance. Hysteresis is a key index characterizing the transmission precision of smallsize joint. Hysteresis measurement and evaluation are the foundation for improving the performance of smallsize joint. Up to now, the hysteresis measurement of smallsize joint is performed mainly based on its main component—reducer. Little research has been done on the overall hysteresis of the smallsize joint. In order to investigate the generation mechanism of total hysteresis of smallsize joint and its basic law, an experiment study on the overall hysteresis of the smallsize joint was conducted based on the developed tester. Based on the analysis on the structural composition of smallsize joint, the composition of total hysteresis and the measurement model, 4 dynamic measurement experiments on smallsize joints A, B and C were conducted. It is found that the geometric hysteresis and elastic hysteresis are the main factors influencing the total hysteresis of the joint. The revolution speed has less influence on the total hysteresis than elastic deformation. Under the conditions of the same load and different revolution speeds, the difference of the measurement results is 005°. It is also found that under certain condition, the dynamic hysteresis and load of the joint show an approximate linear relationship. Finally, a static measurement experiment on the total hysteresis of the joint was conducted, and the result was compared with that for dynamic measurement experiment. It is found that the static measurement result is less than the dynamic measurement result, and the difference is about 01°. The experiment study on the total hysteresis measurement of smallsize joint in this paper lays a solid foundation for establishing the overall quality evaluation system facing to smallsize joints. .txt

Abstract:Abstract:In order to judge the balance between the left and right buckets of the TBR, the difference judgement method and the difference rate judgement method are proposed. The difference judgement method uses the difference between the average water weights of the left and right buckets as the criterion, and let α be the difference limit. The difference rate judgement method uses the ratio of the absolute value of the difference between the left and right buckets to the average of left and right bucket measurements as the judgement criterion, and let β be the difference rate limit. The experiment analysis on the two TBRs of JDZ05 and JDZ10 was carried out. The results show that: 1) For JDZ05, α may take 01 g, β may take 1%; for JDZ10, α may take 02 g, β may take 1%. 2) When the left and right buckets are unbalanced, the rainfall ranges that are not conducive to JDZ05 and JDZ10 monitoring are 0~55 mm and 0~19 mm, respectively; after balancing the left and right buckets, the rainfall ranges unfavorable for JDZ05 and JDZ10 monitoring are 0~35 mm and 0~7 mm, respectively. 3) After balancing the left and right buckets, the difference value, the difference value rate of the left and right buckets and the dispersion of the measurement errors can be reduced, the repeatability of the measurement error can be improved, which is helpful to improve the accuracy of the rainfall measurement. The research results can provide a reference for developing TBR bucket balance judgement method and improving TBR calibration process. .txt

Abstract:Abstract:In order to improve the efficiency and convergence speed of reinforcement learning in multirobot behavior optimal decision making control, the distributed Markov modeling and control strategy for multirobots are studied in this paper. According to the limited perception ability of the robots, an individualcooperative trigger perception function is designed. The individual robot calculates the individualcooperative trigger response probability from the environment observation results, and defines that after a trigger process the joint strategy calculation starts, which reduces the communication amount and computing resources among robots. The Qlearning algorithm is improved through introducing the duallearning rate strategy, which is applied to the behavior decisionmaking of robots. The simulation experiment results show that the algorithm proposed in this paper has quite high cooperative efficiency when the number of robots in the group is about 20. The unit time step ratio is 1085 0. At the same time, the distance adjustment parameter η has an influence on the cooperative search efficiency of the robot. When η is 0008, the required moving time step ratio and average moving distance can reach minimum. Through introducing the double learning rate, the proposed algorithm possesses higher learning efficiency and applicability compared with the reinforcement learning algorithm based on environment model, the average performance improvement reaches about 35%. The proposed algorithm has a high theoretical significance and application value for improving the autonomous cooperative ability of multirobots. .txt

Abstract:Abstract:Geometric parameters modeling is the basis of robot calibration, which directly affects the robot positioning accuracy. The common geometric parameters model has the singularity when the two adjacent axes of the robot are vertical and close to vertical. To solve this problem, a zero reference model (ZRM) based on the direction vector and the connection vector is formulated. It not only meets the requirements of completeness and continuity, but also it is simple and intuitive to calculate the robot endeffector position and orientation. The error model of geometric parameters is established. The endeffector positions and orientations of Staubli Tx60 and ER10LC10 industrial robots are measured by Leica AT960 laser tracker. The reduntant parameters are removed by the orthogonal triangle decomposition method and the geometric parameters errors are identified by the LM algorithm. The results are compared with the output of the MDH model. Experimental results show that the average absolute positioning accuracy of robot endeffector cabibrated by ZRM model is improved by 75%~90%, which is obviously higher than that cabibrated by MDH model. This method is suitable for industrial robots with high positioning accuracy requirements. .txt

Abstract:Abstract:Dental caries, as one of the three major prevention and treatment diseases of human, seriously affects people′s oral health. Tooth restoration is an important method for treating dental caries. Tooth preparation is a necessary treatment link in the oral dental caries restoration process. In traditional tooth preparation process, a large number of repetitive fine adjustments are required, which rely on the manual operations of doctors with rich clinical experience. In this paper, a robot is adopted to assist doctors for tooth preparation, which effectively improves the quality of tooth preparation and the effect of oral treatment. Through analyzing the operating characteristics of the doctor′s preparation process, the specific requirements of the posterior tooth full crown preparation are determined, and the design requirements of the tooth preparation robot are specified. The constraint conditions of surface patches are constructed using the curved surface mapping relationship, then the model points on the curved surface of the preparation tooth are determined. The functional cusp bevel is divided into three curved surfaces according to the adjacent relationship between the teeth and functions. The interpolation of the three curved surfaces is completed based on the NURBS curved surface. According to the curved model points and characteristics of the curved surface patches, the tooth preparation trajectory planning of the tooth preparation robot is realized based on the isoparametric method. The virtual prototype and physical prototype of tooth preparation robot were designed and developed, and the experiment study on tooth preparation was carried out. The maximum relative fixedpoint errors of the feature points in the X, Y and Z directions are 023, 015 and 097 mm, respectively, and the confidence intervals of the system errors are all within 03 mm, stably. The correlation of the random errors of the feature points in different directions is weak, which verifies the correctness of the tooth preparation trajectory planning method and the feasibility of robot tooth preparation. .txt

Abstract:Abstract:Comprehensive and reliable evaluation of the realtime performance of leg motion intention recognition algorithms is the premise to realize flexible and stable control of lower limb prosthesis. In this paper, a multilayer real time test evaluation method for lower limb motion intention recognition algorithms is proposed, which comprehensively evaluate the reliability, stability and motion intention recognition speed of the algorithms. Using the developed test evaluation system for lower limb motion intention recognition algorithms, two motion intention recognition algorithms based on myoelectric signal source and mechanical signal source were tested in real time, respectively. The results show that the motion recognition time of the myoelectric signal source based algorithm is longer than that of the mechanical signal source based algorithm; however, the stability of the myoelectric signal source based algorithm is better than that of the mechanical signal source based algorithm. Additionally, the performances of normal and abnormal recognition strategies can be effectively distinguished using the proposed test evaluation system, and it is found that the motion recognition stability index for normal strategy is 25% higher than that for the abnormal strategy. These results demonstrate that the proposed real time test evaluation method for lower limb motion intention recognition algorithms can effectively evaluate the real time performance of different signal source based algorithms and different recognition strategies, and can provide a testing platform for the development of intelligent lower limb prostheses control system. .txt

Abstract:Abstract:The service robot is faced with unstructured scene in the task of grasp. Because of the irregular placement and shape of the objects, it is difficult to accurately calculate the robot′s grasp posture. Aiming at this problem, a robot optimal grasp posture detection algorithm with dual network architecture is proposed. Firstly, the YOLO V3 target detection model is improved, which improves the detection speed of the model and the recognition performance of small target objects. Secondly, convolutional neural network is used to design multitarget grasp detection network, which generates the robot grasp area in the image. In order to calculate the optimal grasp posture of the robot, the IOU area evaluation algorithm is established, which screens out the optimal grasp area of the target object. The experiment results show that the target detection accuracy of improved YOLO V3 reaches 91%, and the detection accuracy of the multitarget grasp reaches 86%, the detection accuracy of the robot optimal grasp posture reaches above 90%. In summary, the proposed method can efficiently and accurately calculate the optimal grasp area of the target object to meet the requirements of the grasp task. .txt

Abstract:Abstract:Aiming at the problems of the timevarying load and model uncertainties in industrial robot joint servo system, a variablegain adaptive disturbance rejection control strategy based on inertia estimation is proposed in this paper. First of all, the mathematical model of joint servo system is established. The second order state equation of joint servo system is obtained through frequency domain analysis. In order to reduce the influence of disturbance and uncertain parameters, the linear extended state observer is designed and the adaptive method is used to estimate the moment of inertia. At the same time, robust and sliding mode control are combined to keep the system stability. The relevant simulation and experiment research on the proposed control strategy were carried out. The experiment results show under this control strategy, the tracking error of the sinusoidal signal at the motor end is less than 02 rad, and the position error under load disturbance is less than 003 rad, which is reduced by about 40% compared with that for single adaptive disturbance rejection control. This controller has strong antiinterference performance, which improves the control accuracy and dynamic performance of joint servo system. .txt

Abstract:Abstract:It is necessary to maintain the observation and detection formation deployed at the position and complete the task with less energy consumption when the masterslave unmanned underwater vehicle (UUV) cluster performs the collaborative observation and detection task. Firstly, the requirements of UUV lowenergy green dynamic positioning and formation maintenance are analyzed. Secondly, under the dominance of the UUV green dynamic positioning strategy, two control schemes of the global optimal thrust planning based on particle swarm optimization and realtime rolling optimization based on the improved generalized predictive control are specifically designed. Compared with the traditional PID control, simulation results show that the control strategy and control method can reduce the control time from 200 s to 60 s. And the dynamic control task can be completed with lower energy consumption on the premise of meeting the requirements of UUV cluster collaborative view detection task. .txt

Abstract:Abstract: How to isolate the pure cell samples with certain characteristics from complex heterogeneous and numerous cells is a difficult problem. It perplexes researchers in life science and clinical pathology for a long time. The methods for effectively isolating pure, undamaged, and highspecific cells have significant influences on the explanation of life process, the study of pathological mechanism, the determination of treatment plan, and the analysis of drug efficacy. In this study, the research progress of single cell isolation methods and instrument is reviewed and summarized. Fluorescent activated cell sorting, microfluidics, laser microdissection and micromanipulator are emphasized. Firstly, for the cell isolation methods, the working principle and the corresponding theoretical models are presented. Secondly, the cell isolation instrument, the structural feature, research progress and main performance parameters are analyzed. Finally, the advantages and limitations of various single cell isolation methods are discussed. Meanwhile, the trend of single cell isolation methods is prospected. .txt

Abstract:Abstract:There are significant individual differences of steadystate visual evoked potential (SSVEP) responses and the different quality of EEG signals collected by different electrode channels. To solve these problems, an adaptive method of synthesizing frequency response characteristics and weight coefficient is proposed and verified by experiments. First, four subjects perform three SSVEP frequency sweep experiments. SSVEP amplitudefrequency characteristic response passband of eight electrodes in the cerebral occipital region is achieved. Secondly, according to the average signaltonoise ratio of the electrode channels, the weight coefficient of each electrode is received. Then, the subject′s amplitudefrequency characteristic response passband is obtained. Finally, to avoid the intense visual fatigue caused by lowfrequency flicker, the midband (15~30 Hz) of the individual′s amplitudefrequency characteristic response passband is selected as the stimulation frequency for braincomputer interface experiments. Experimental results show that the proposed adaptive steadystate visual evoked braincomputer interface has high accuracy (9709 % on average) and information transmission rate (10026 bits/min on average) when the recognition time is 3 s. The visual fatigue is effectively reduced. Research results provide new ideas for the design of a BCI based on individual differences. .txt

Abstract:Abstract:Focal EEG recognition can provide important reference value for epilepsy surgery. This paper proposes a focal EEG recognition algorithm based on deep network with transfer learning. Firstly, the continuous wavelet transform (CWT) is used to perform timefrequency analysis on the EEG signals and obtain the timefrequency map of the EEG signals. Then transfer learning is performed on the AlexNet model, and the network structure is adjusted to adapt to focal EEG recognition. The output of the seventh fully connected layer of the model is used as the characteristic presentation of the timefrequency images. Finally, the classification algorithms of SVM, BP, LSTM, SRC and LDA are used to classify the features. In this paper, based on the open source EEG dataset, the 10fold crossvalidation algorithm is adopted to verify the algorithm, the effects of the six classifiers are compared. The average specificity, sensitivity and accuracy of the SVM algorithm are 8807%, 8881% and 8844%, respectively, which proves the effectiveness of the method in focal EEG recognition. .txt

Abstract:Abstract:It needs to control ventilator parameters according to individual differences of patient in the auxiliary treatment of mechanical ventilation. In this study, the mechanical model of a respiration system based on general regression neural network（GRNN）are analyzed. To identify parameters of the respiratory system model, a fusion method based on PSO_GRNN, numerical integration and recursive least square is proposed. The static lung pressure value of singlecycle respiratory samples is calculated by direct calculation and the second order polynomial is used to fit the volume error. The mean absolute error of static data points for ten inhalation cycles is 0169 3 mL, and the mean absolute error of static data points for ten expiratory cycles is 0372 8 mL. PSO_GRNN is used to predict the static lung pressure of the multicycle respiratory sample set. For the ten sample sets of respiratory cycle, the average error of the training set is 0009 1 and the average error of the test set is 0406 5. Simulation results show that PSO_GRNN is better than PSO_BP in terms of convergence rate, average error and computation speed. The proposed method can provide an effective reference basis for doctors to set ventilator parameters during the mechanical ventilation treatment. .txt

Abstract:Abstract:At present, there is a safety problem of specific absorption rate (SAR) in the resonant wireless charging system of pacemaker. To avoid the damage caused by SAR, a resonant wireless energy supply system based on LCCC compensation is designed in the frequency range of 10～100 kHz. The influences of frequency, coil offset and compensation parameters on the transmission performance of the system are considered. A loss circuit model is formulated, which considers the equivalent series resistance. By analyzing the output power and transmission efficiency of the system under different frequencies, LC parameters and coupling coefficients, the optimal frequency and compensation parameters are comprehensively determined. Finally, an experimental system is set up to check the system′s energy supply efficiency, antioffset capability, and corresponding safety indicators. Results show that the transmission efficiency of the system can reach 433%～732% and the maximum temperature rise is only 08℃ under the condition of 0～2/3 lateral offset of the center distance of two coils. Meanwhile, SAR evaluation is not necessary. It provides an effective method for improving the safety design of human implantable equipment. .txt

Abstract:Abstract:To improve the image resolution in magnetic detection electrical impedance tomography (MDEIT), an algorithm based on structural prior information is proposed. The preprocessed CT image is segmented by using the level set method to obtain the structural information image. The lung edema lesion simulation model is formulated by CT image, and the traditional sensitivity matrix algorithm is utilized to reconstruct the MDEIT reconstructed image. Experimental results show that the MDEIT algorithm based on structural prior information can reduce the artifacts of reconstructed images. The relative image error is reduced from 8847% to 3339%. And the image correlation coefficient is increased from 033 to 081. Compared with the traditional reconstruction algorithm, the accuracy of the reconstructed image structure and the reconstruction accuracy of the conductivity based on structural prior information are improved. The proposed method algorithm provides a foundation for the clinical application of MDEIT.

Abstract:Abstract:The phase locked stimulus technology has great application prospect in the neural mechanism research and clinical treatment. However, the problem of phase locked between electroencephalogram (EEG) and stimulus need to be solved. Due to the complicated timevarying characteristics of EEG, there is still lack of effective stimulus algorithms that can be used to lock with the EEG phase. Therefore, a phase locked stimulus method for EEG is proposed, which is based on the variational mode decomposition (VMD) and autoregressive (AR) prediction. Firstly, EEG is processed by VMD to obtain multiple eigenmode signals. Then, each eigenmode signal is predicted by the AR model. The predicted values corresponding to all modes are accumulated. Finally, according to the frequency and phase characteristics of the accumulated results, the stimulus is generated, which is phaselocked with EEG. The method is evaluated in the synthesized EEG and 20 subjects (aged 20~36, male 12, female 8) offline resting EEG respectively. Results show that VMDAR can overcome the influence of EEG instability and generate the stimulus with higher phaselocked value (PLV). When the length of prediction time increases from 001 s to 04 s, PLV of opened EEG decreases from 099 to 039, and PLV of closed EEG decreases from 099 to 065. When the length of modeling time increases from 025 s to 25 s, PLV of opened EEG increases from 064 to 083, and PLV of closed EEG increases from 053 to 065. The phase locked performance of VMDAR is superior to the methods of AR and AR based on empirical mode decomposition EMDAR under all test conditions. This method can also be applied to other nonstationary closedloop phaselocked systems. .txt

Abstract:Abstract:Aiming at the low accuracy problem in the data fusion of indoor pedestrian positioning and navigation using traditional Kalman filter algorithm, this paper proposes a new pedestrian dead reckoning navigation algorithm using rank Kalman filter (RKF) based on pedestrian dead reckoning technology. RKF technology can nicely handle nonGaussian and nonlinear system due to its special rank sampling mechanism. Through combining RKF technology and zero velocity update(ZUPT) technology, the algorithm can fuse the multisensor data measured in indoor pedestrian motion, and achieve more accurate indoor pedestrian positioning and navigation. Firstly, the zero velocity detection algorithm is used to analyze and obtain the zero velocity information from the data measured with MEMS sensors. Then, the obtained zero velocity information is used as ZUPT, which is fused with the information from the RKF algorithm and the pedestrian position is obtained. Finally, the experiment result shows that indoor pedestrian dead reckoning(PDR) algorithm based on RKF achieves certain improvement compared with the pedestrian dead reckoning algorithm using extended Kalman filter, and reduces the indoor pedestrian navigation and positioning error by about 1891%. .txt

Abstract:Abstract:In order to solve the problem that the curvature radius of the concave and convex testing surfaces of a curved surface component influences the resolution/sensitivity and positioning/quantitative deviation of the fault detection of an angled shearvertical (SV) wave electromagnetic acoustic transducer (EMAT), a finite element model for the radiated sound field of an angled SV wave EMAT based on curved testing surfaces (concave and convex surfaces) is established. The influences of the number of the turns of the meanderline coil and the curvature radius on the main lobe peak, main lobe width, maintosidelobe ratio and incident angle of angled SV wave are investigated, and the simulated results are verified by experiments. The results show that when the numbers of the turns of the meanderline coil are the same, the SV wave amplitude of the convex testing surface is improved by more than 348% and the main lobe width is reduced by 438% compared with those of the plane testing surface. When the number of the turns of the meanderline coil increases from 10 to 30, the SV wave amplitude of the convex testing surface is improved by 423% and the main lobe width is reduced by 436%. The sound field characteristic of the convex testing surface is better than those of the plane and concave testing surfaces. The curvature radius of the convex testing surface has a significant influence on the main lobe peak, main lobe width and the maintosidelobe ratio of angled SV wave. .txt

Abstract:Abstract:The realtime working conditions of the ball mill during the grinding process are complicated. It is difficult to accurately obtain the internal load status of the ball mill. In this study, the energy difference between the original cylinder vibration signal and the intrinsic mode function is utilized as the evaluation parameter of the adaptive variational mode decomposition (VMD) layer number. In this way, a new autocorrelation function is formulated. The intrinsic mode function is processed by introducing the RifeVincent selfconvolution window and the energy centrobaric method. A feature extraction method for ball mill load based on the adaptive VMD and the improved power spectrum estimation is proposed. The ball mill load identification system based on LabVIEW is developed. The number of layers of intrinsic mode function can be adaptively determined. The algorithm's abilities to resist modal aliasing and false components are enhanced. The accuracy of ball mill load detection is improved. Measurement results show that internal load features of ball mill during grinding process are effectively extracted, and the mill load status is accurately identified. This method provides accurate and reliable basis for the optimization control and efficiency improvement of the grinding. .txt

Abstract:Abstract:In order to realize the fault warning and identification of front bearing of wind turbine generator, in this paper a timefrequency domain modeling method is proposed, which integrates the time series data of supervisory control and data acquisition (SCADA) system with the vibration data of condition monitoring system (CMS). Firstly, the temperature model of generator front bearing based on gated recurrent unit (GRU) neural network is established using the SCADA data, and the temperature residual features are calculated. Secondly, the time domain features and frequency domain features of the vibration signal of generator front bearing are extracted. Finally, the temperature residual features and the timefrequency domain features of the vibration signal are fused, and the extreme gradient boosting (XGBoost) based fault identification model of the front bearing is established, which can identify five working conditions of the generator front bearing, including normal, inner ring damage, outer ring damage, shaft imbalance and rolling body damage. Extensive experiment results demonstrate that the proposed method can achieve higher identification accuracy compared with the front bearing fault warning identification method using the vibration signal characteristics alone. The average identification accuracy for normal, inner ring damage and outer ring damage conditions increase from 87%, 585% and 65% to 885%, 675% and 74%, respectively. .txt

Abstract:Abstract:Aiming at the problem that traditional cyclostationary signal processing method based on second order statistics can not effectively deal with impulsive noise interference, a new fault detection method based on cyclic multiple kernel correntropy is proposed. Firstly, the definition of the multiple kernel correntropy is given, the calculation formulas of the cyclic multiple kernel correntropy function and the cyclic multiple kernel correntropy spectral density function are deduced. The denoising mechanism of the cyclic multiple kernel correntropy is analyzed. Secondly, the denoising performance of the cyclic multiple kernel correntropy under lower signal to noise ratio is verified with simulative signal, the result shows that the cyclic multiple kernel correntropy can not only effectively suppress the Gaussian noise, but also effectively suppress the nonGaussian noise. The cyclic multiple kernel correntropy provides a robust processing method for Gaussian and nonGaussian noise. Finally, the cyclic multiple kernel correntropy method was applied to the fault diagnosis of gearbox tooth wear. The experiment results show that the cyclic multiple kernel correntropy has demodulation function, can accurately characterize the spectral characteristics of gear tooth wear fault, and effectively extract the weak signal submerged in strong noise environment. The cyclic multiple kernel correntropy method is proved to be an effective method for gear fault diagnosis. .txt

Abstract:Abstract:It is difficult to detect the simultaneous multileak of the pipeline effectively. To solve this problem, a multileak diagnosis algorithm based on the fault diagnosis observer is proposed. Firstly, the residual generator is formulated based on the fault diagnosis observer and the leak coefficients are extracted by the residual error. Secondly, the disturbance decoupling technique is used to make the observer robust to unknown disturbances. The stability of the fault diagnosis observer is proved by using the Lyapunov function and Barbalat lemma. The approach can detect and estimate the leak coefficients. Simulation results of the multileak experiment prove that the maximum error of leak coefficients is 1×10-5 m5·s-1, and the minimum error of leak coefficients is 02×10-5 m5·s-1. Field experiments evaluate the feasibility and effectiveness of the proposed algorithm. .txt