Abstract:Spacecrafts for earth and celestial observation play a key role in important fields related to national economic and social development, such as deep space and polar exploration, military investigation, surveying and mapping of land and resources, geological and agricultural disaster monitoring, marine observation, meteorological observation and environmental monitoring. The on-orbit structure of the remote sensing spacecraft may undergo micro deformation because of extreme temperature change in space, gravity, debris impact, fatigue, etc. The on-orbit micro deformation monitoring plays an important role in ensuring the realization of functional indicators of remote sensing spacecraft and the reliability and service life of the system. Optical fiber monitoring technology is regarded as one of the most potential on-orbit monitoring technologies for remote sensing spacecraft. After research and development over the past three decades, it is evolving from laboratory research to engineering application. However, there are some problems that limit the application and development of this technology, which need to be solved urgently. Therefore, this article reviews the main technologies for monitoring the micro deformation of the structure of remote sensing spacecraft, analyzes the main technical types, technical advantages and disadvantages, micro deformation field reconstruction algorithm, on-orbit applicability, and typical application cases of optical fiber monitoring, and illustrates the key problems to be studied and solved and the future development direction.

Abstract:The oil and gas pipeline sizing technology is an important tool for detecting changes in the pipe diameter and ensuring pipeline safety. To address the problems that the traditional thin aluminum discs cannot detect multiple defects and cannot be reused, this article designs a flexible sizing sensor, which measures the strain change when the resistance is encountered through the resistive strain gauges placed in the flexible polyurethane sizing discs to achieve quantitative detection of defects in the pipe. The coupling model between the strain gauges and the polyurethane sealing disk is formulated with ABAQUS software, and the strain output characteristics are analyzed when tensile and compressive deformations are generated. The function of the flexible sizing sensor is tested by a flexible sizing sensor performance testing experiment bench. Finally, the BP neural network and the genetic algorithm are combined to derive the structural parameters that make the sensor detection performance optimal. For the DN200 pipeline, research results show that when the outside diameter of the sizing disc is 97. 5% of the inner diameter of the pipe, the thickness of the sizing disc is 19 mm, and the sizing disc flap spacing is 8 mm. The flexible sizing sensor can better characterize the size of the depression, which has certain engineering application value for oil and gas pipeline deformation detection operations.

Abstract:To meet the requirements of the multi-phase flow water holdup measurement at the surface wellhead of oil production, a permanently dynamic and static water holdup measuring device for surface wellhead (PDSWHMD_SM) is designed and developed by combining dynamic conductance measurement and cylinder capacitance static measurement techniques. This article establishes a numerical model of the permanent conductance-capacitance integrated sensor (CCIS) using the finite element method (FEM). And the medium distribution characteristics of the fluid flow and stationary state in the pipe, CCIS structural parameters, electric field distribution, and response characteristics have been studied in depth. The optimal structure parameters of CCIS are finally determined: He = 90 mm, ID= 30 mm, Ihe = 3 mm, Hc = 375 mm, IR1 = 26 mm, Tc = 1 mm, Hm = 56 mm, which are proved to be within 5% of the measurement error. In addition, this article conducts an experimental study with multi-phase flow conditions such as the total flow rate range of 5~ 70 m 3 / d and the water holdup range of 50% ~ 90% . The results show that the developed device PDSWHMD_SM has good water holdup measurement performance with the error of water holdup measurement within 5% . Both simulation and experiment demonstrate that PDSWHMD_SM performs well in water holdup measurement.

Abstract:Current intensity is the basic parameter of the electrolytic aluminum process. The accurate measurement of electrolytic cell current is beneficial to improve the production efficiency and work stability. This article analyzes in principle about the reason why the fiber-optic current sensor has high accuracy in electrolytic aluminum current measurement compared with the traditional current measurement technology. To reduce the repeated disassembly and assembly error and high current temperature error of fiber-optic current sensor in field applications, innovatively proposed key technologies such as the sensing optical fiber plug-in scheme based on flexible packaging, two-dimensional compensation of temperature error and nonlinear error, etc. The handheld fiber-optic current sensor is designed and developed for the first time. The test results show that the error of the sensor under repeated disassembly and assembly is less than 0. 12% , the error when measuring 0. 5 ~ 30 kA current in the temperature range of -40℃ ~ 70℃ is less than 0. 2% . When measuring the column bus current in an electrolytic cell, the error with the displayed value in the control room is less than 0. 4% . Through the continuous monitoring and analysis of the distributed current of the cathode soft discharge, the early warning of the damage of the electrolytic cell is realized.

Abstract:The wear particle monitoring technology has become a key part of the intelligent diagnosis of aero-engine, because the wear of rotating parts of aero-engine is an important factor that affects the lifespan of aero-engine. The parameters of the wear particle in the lubricating oil can directly reflect the wear state of rotating parts. In this article, a comb type capacitance sensor is proposed, which uses the structure of built-in multi-parallel plate electrodes. On the basis of analyzing the wear particle detection principle of the sensor, we determine the structural optimization direction and the optimal parameter through numerical simulation. Experimental results show that when the diameter of iron wear particles increases from 0. 3 mm to 0. 9 mm, the variation in capacitance increases from 3 Ff to 17 Ff. When the number of wear particles increases from 1 to 4, the variation in capacitance is approximately in a positive linear relationship with the particle number, and the maximum increase ratio is about 200% to 350% . When the power frequency increases from 10 kHz to 100 kHz, the drop ratio of variation in capacitance is 36% to 67% . The results show that the comb type capacitive sensor can improve the uniformity of the internal electric field, which could improve the consistency of the results and the detection rate of wear particles, and detect 200 μm diameter particles in 10 mm diameter pipe which shows a good discrimination in outcome. This technology can provide technical support for the monitoring research of aero-engine oil wear particles.

Abstract:To meet the urgent demand for precise two-dimensional positioning technology in the fields of semiconductor industry and aerospace, a planar two-dimensional time grating displacement sensor based on the orthogonal double traveling wave magnetic field is proposed in this article. The sensor consists of a stationary part and a movable part. The stationary part is composed of a magnetic conductive base and the excitation coils arranged in the x and y directions. The movable part is composed of the magnetic conductive base and the induction coils arranged in the x and y directions. An orthogonal traveling wave magnetic field is generated when the sine and cosine excitation signals are applied to excitation coils. The displacement values in x and y directions are obtained by calculating the inductive electromotive force. In this article, the working principle and structure of the sensor is firstly introduced. Then, the electromagnetic field simulation of the sensor model is carried out. According to the simulation results, the error is traceability analyzed and the sensor structure is optimized. Finally, the sensor prototype is fabricated by the printed circuit board technology, and the corresponding electrical system is designed for experimental verification. Results show that the sensor can realize two-dimensional displacement measurement, and the measurement range is up to 160 mm×160 mm. The peak to peak value of displacement error in the x-direction is 32. 8 μm, and the peak to peak value of displacement error in the y-direction is 34. 5 μm within a pitch.

Abstract:A design and fabrication method of miniature optical fiber Fabry-Perot pressure sensor based on hyper-elastic material is proposed to meet the needs of vivo interventional medical applications. Through theoretical analysis, a Mooney-Rivlin mechanical simulation model suitable for hyper-elastic silicone rubber material is formulated. The compression deformation state of pressure sensing materials with different components and thickness is theoretically analyzed, and the optimized sensor materials and structural parameters are obtained. The fabrication method of the sensor is further proposed, and the pressure-sensing performance of the sensor is evaluated by the pressure-sensing performance test, temperature influence test and the in vitro blood pressure test. The results show that when the diameter of the pressure-sensing material is 180 μm and the thickness is 250 μm, the pressure sensitivity of the sensor within the pressure measuring range of 0 ~ 40 kPa reaches 154. 56 nm/ kPa, and the relative error of pressure measurement caused by a large temperature range of 20℃ ~50℃ is only 0. 36% . The influence of temperature on pressure measurement is completely negligible. Compared with the traditional diaphragm optical fiber pressure sensor, the miniature optical fiber Fabry-Perot pressure sensor based on the hyper-elastic material is not only small in size, high in sensitivity, but also has the technical advantages of low cost and convenient fabrication.

Abstract:The eccentricity error is one of the main sources of circular grating angle sensors′ measurement error. By deriving the expression of eccentricity error and its trigonometric series, it demonstrates that the eccentricity parameters can be measured and the eccentricity error can be compensated by using two reading heads in opposite direction. Therefore, a procedure to obtain the eccentricity parameters from two reading heads′ counts data by the improved particle swarm optimism algorithm is raised and a real-time FPGA eccentricity error compensation module based on the interval transform and two-tier look-up table is designed, which makes it possible to use only a single reading head after eccentricity parameters have been obtained and achieve virtually the same accuracy as the two reading heads method. The experimental data show that, for the encoder of 320 000 pulses per race used in the experiment, the maximum difference between the counts of a single reading head and the mean counts of two reading heads is 109 before error compensation. Thus, the eccentricity error of a single reading head can be up to 0. 06°, which has a non-negligible influence on angle measurement accuracy. After error compensation, the maximum difference between these two values is 6, and the average difference is only 1. 46, which evaluates the effectiveness of the error compensation method proposed in this article.

Abstract:Based on matter wave and Sagnac effect, the angular velocity and acceleration in horizontal direction of the carrier can be measured by the horizontal atom interferometric gyroscope. The physical system is the key subsystem of a gyroscope, which provides the circumstances of ultrahigh vacuum with special magnetic distributions and optical paths for control and detection of atom states by lasers. To realize of atom interferometry and satisfy the requirements of high-precision measurement, the principles of atom interferometric gyroscope operation are analyzed, and the functional requirements and basic mechanical components are clarified. Based on the theoretical analysis, the key specifications including vacuum degrees and magnetic distributions of each functional area are proposed. Therefore, a physical system of atom interferometric gyroscope is constructed, the measurement results reveal that the vacuum held steady at 10 -8 Pa, and the diameter of cold atom cloud generating from this system is 6 mm, with the quantity of 8×10 9 and temperature of 12. 8 μK, which satisfies the requirement of the subsequent experiments.

Abstract:To address the problem of accuracy loss of embedded angular displacement sensor after long time use, a self-calibration method using single probe error phase shift method is proposed in combination with the structural characteristics of the sensor. Firstly, the sensor error characteristics are analyzed based on the Fourier transform according to the principle of circumferential spatial closure. Secondly, the sensor error model is constructed by transforming the error sequence and solving the function relationship between the error and the starting reference with the sequence of starting measurement values as the starting reference for error offset. Finally, the online calibration experiment platform is established for verification experiments. Experimental results show that the single intra-pole error results are consistent with the measurement error characteristics, and the whole cycle measurement error is reduced significantly, with the error peak value reduced from 127. 80″ to 5. 90″. Comparison experiments and repeatability experiments show that the calibration efficiency of the method is improved by 80. 69% compared with the comparative calibration method, and the error distribution is concentrated after calibration, while remains good measurement stability. The self-calibration method described in this article effectively solves the problem of sensor accuracy loss without relying on a high-precision reference apparatus, and has important application significance for realizing the self-calibration of embedded angular displacement sensors.

Abstract:The double roller-guide involute measuring instrument is a kind of involute measuring instrument with no abbey error, few error sources, and high measurement accuracy, which is commonly used to measure the class-1 gear involute artifacts or class-1 standard gears. The effect of ambient temperature on the double roller-guide involute measuring instrument and gear involute artifact is analyzed. The change of the ambient temperature would not bring about the profile slope deviation as the thermal expansion coefficient of the roller of the double roller-guide involute measuring instrument and the gear involute artifact to be measured are the same. However, when the thermal expansion coefficient of the roller and the gear involute artifact is different, the profile slope deviation under the change of the ambient temperature within 20℃ ± 5℃ is linear with the ambient temperature. The variation of ambient temperature only causes nanometer-scale profile form deviaiton in both case and can be neglected. The thermal expansion simulation of the gear involute artifact, roller, and the guide is verified by ANSYS. The measurement experiments for profile deviation at different ambient temperatures are implemented on the same tooth surface of the gear involute artifact made of GCr15 steel. As the ambient temperature changes, the profile slope deviation measured by the roller made of GCr15 steel does not change significantly, and the change of profile slope deviation measured by the roller made of SiC ceramic material is 0. 71 μm/ ℃ , which is basically consistent with the theoretical value of 0. 73 μm/ ℃ , and the tooth profile form deviation measured by the roller of both materials does not change significantly.

Abstract:In this article, the data synchronization problem of dynamic comparison between turntable angular measurement and ring laser gyroscope is studied. The synchronization problem between the turntable angular measurement and the ring laser gyroscope is analyzed and the synchronous angle deviation introduced by their different system delay is elaborated. Based on the analysis of synchronous angle deviation, a delay compensation scheme is proposed to make the output value of turntable angular measurement and ring laser gyroscope to the same turntable angle position at the same time. A real-time compensation circuit for synchronous angular deviation based on field programmable gate array is developed, and the key links of the circuit in the process of compensation are analyzed in detail. The experimental platform of dynamic angle measurement is established to evaluate the compensation effect of synchronous angle deviation and the results show that the proposed method can achieve outstanding compensation effect at different speeds. The difference of compensation effect at each speed is less than 0. 5″.

Abstract:The accuracy design is an effective method to improve machining accuracy of machine tools. How to identify key geometric errors accurately and assign weights to each geometric error are the prerequisites for achieving accuracy design. Therefore, a sensitivity analysis method for identifying the key geometric errors is proposed. Firstly, the machining error model of machine tools based on the multi-body system theory is formulated. And the sensitivity analysis model based on the machining error model of machine tools is established. Meanwhile, a new sensitivity index is defined. Then, the 10th, 17th, 22nd, 24th and 37th geometric errors are determined as the key geometrical errors of the machine tool by simulation analysis. Meanwhile, the weight of each geometric error is allocated. Finally, the “S” shape test piece is selected as the research object for experiment verification. The “S” shaped test pieces are machined by compensating the identified key geometric errors and all geometric errors. And the form errors of the “ S” shape test piece are compared. The comparison results show that the difference of the average profile error obtained by two compensation strategies on three detection lines is very small, which is 0. 005, 0. 004, and 0. 006 mm, respectively. Therefore, the correctness of sensitivity analysis method is verified.

Abstract:This article proposes a new method for sensitivity analysis of machine tool position errors. Firstly, the position error model of five axis gantry machine tool is formulated, which is based on the multi-body theory and homogeneous transformation matrix. Secondly, the geometric error parameters related to the position are characterized by the truncated Fourier technique. The sensitive value of each error parameter to position error can be expressed as its Fourier amplitude square. After normalization, the key geometric error parameters are the 2nd, 3rd, 8th, 15th and 26th errors. Compared with the traditional Sobol method, the simulation results show that the key geometric errors identified by the two sensitivity analysis methods are the same and the sensitivity values are similar. In addition, the calculation efficiency of sensitivity analysis of the proposed method is better than that of traditional Sobol method. Finally, to verify the validity of key geometric errors, a compensation experiment on key geometric errors of machine tools is proposed. Experimental results show that the machining accuracy of the machine tool is improved by 48% after the key geometric errors are compensated. Therefore, the sensitivity analysis method of machine tool position error in this article is feasible and effective.

Abstract:To address the problems of poor positioning accuracy and mapping error in the traditional simultaneous localization and mapping algorithms under the complex dynamic environments with dynamic objects, a semantic RGBD-SLAM algorithm in dynamic scenes is proposed, which is based on the optical flow. Firstly, the camera ego-motion is compensated by the optimized 2D perspective correction method based on adjacent frames. Secondly, by feeding the compensated perspective images into the RIFT-S network, the lowresolution dense optical flow field is obtained for extracting the current mask of the dynamic region. The dynamic regions in the current mask are tracked and optimized by using the position and velocity of the dynamic regions in previous mask. The accurate dynamic regions in each frame can be extracted. Finally, the static and dynamic features are separated, and the optimized camera pose is obtained by minimizing the reprojection error of the static feature points. The static semantic octree map without people is established by the depth data from camera and semantic information produced by the lightweight semantic segmentation network BiSeNetv2. Compared with ORBSLAM2, the test results on the public data set of TUM indicate that the absolute trajectory error of the proposed algorithm is reduced by more than 90% , and the accurate masks of dynamic regions and an accurate semantic map also can be obtained. Results show that the proposed algorithm has a good positioning accuracy and robustness under complex dynamic scenes.

Abstract:The electrocardiogram (ECG) reflects the cardiac electrophysiological activity, which is essential for the automatic diagnosis of heart disease. In this article, an automatic classification method of arrhythmia using multi-scale recurrence plot and vision transformer is proposed. Firstly, the ECG is decomposed into low-frequency components and several high-frequency components using wavelet transform, which are respectively transformed into 2D texture image by the recurrence plot method. To solve the problem of sample imbalance, the conventional vision transformer is improved by replacing cross entropy loss with multi-classification focal loss. Finally, the arrhythmia classification is performed by utilizing the multi-scale recurrence plot representation of ECG and the improved vision transformer. The MIT-BIH arrhythmia dataset is utilized to evaluate the proposed arrhythmia classification method. The average accuracy of the proposed method is 97. 38% . Experiment results show that the proposed method is effective and better than other conventional method.

Abstract:Aiming at the problem that the existing three-dimensional point cloud imaging methods for millimeter-wave radar are difficult to give consideration to both high resolution performance and computing speed, a fast high resolution three-dimensional point cloud imaging method combining adaptive grid evolution and SAMV is presented. Based on the range image obtained by FFT, the range bin of the target detected by CFAR is used to estimate the distance of the target. Then, the adaptive mesh evolution method of dividing coarse mesh-power spectrum estimation-detection-mesh refinement is used for the azimuth-elevation angle of each range bin with the target to obtain the refined mesh of interest. After estimating the power spectrum for these thinned grids, the detection is performed again to quickly obtain the three-dimensional point cloud image of the target. The power spectrum of the grid is estimated by using 2D SAMV before and after grid splitting, which improves the information accuracy and resolution. The results show that the proposed method can increase the speed of generating three-dimensional point clouds to about 10 times that of the 2D SAMV method while maintaining the angular resolution of 4 degrees.

Abstract:To solve the problems of difficult pose measurement of tunneling equipment in low illumination, high dust and multi-light environment in coal mine, a visual positioning method of tunneling equipment based on the improved RANSAC feature extraction is proposed. Firstly, the three laser target images collected by the mine explosion-proof camera are preprocessed, and the shape and wire frame models are established respectively. Then, according to the shape model, the coordinate extremum is taken as the initial point of the elliptic model. The ratio of the difference between the previous and the latter two internal points is the optimal number of iterations, and the optimal parameters of the elliptic model are iteratively obtained to extract the point feature. According to the wireframe model, the pixel coordinate model is used as the initial point of the straight line model, and the line features are obtained by using the adaptive condition threshold and the sampling times. Finally, the point line feature is used as the input of the 3P3L pose solution model, and the pose information of the tunneling equipment is obtained by spatial coordinate transformation. Experimental results show that the relative error of the visual positioning method described in this artcle is ±45 mm within the range of 80 m from the three laser target, which can basically meet the positioning requirements of the coal mine tunneling equipment. It provides a new idea for the pose measurement of the tunneling equipment in the harsh environment of the coal mine.

Abstract:For the imaging of plate-like objects, the computed laminography ( CL) technology has been widely used because of its nondestructive and intuitive characteristics. CL imaging methods are mainly divided into linear scanning and circular scanning. Linear scanning has high efficiency, and circular scanning has good data uniformity. However, due to the incomplete projection data of both methods, there are problems of sliced image aliasing and unclear image edge. In this article, a hybrid scanning CL (HyCL) imaging method is proposed, which consists of linear scanning and circular scanning. The geometric model is formulated, the simulation and imaging experiments are implemented by using SIRT image reconstruction algorithm, the imaging effect and in-slice aliasing is analyzed. Compared with parallel translation computed laminography ( PTCL), orthogonal translation computed laminography ( OTCL) and rotational computed laminography (RCL), results show that HyCL can effectively reduce aliasing artifacts and protect details. The image contour can be retained clearly and evenly, and the contrast is better.

Abstract:Vision multi-robot cooperative SLAM mainly uses cameras as sensors and achieves localization and map building through multi-robot cooperation. However, the front-end computation is too large in the face of complex environments, which tends to lead to unsatisfactory overall system accuracy. Inspired by the lightweight features of REVO and SVO algorithms, this article proposes a multirobot cooperative SLAM system based on lightweight intelligence, aiming to reduce front-end computational resources while improving system scalability. This article proposes an improved REVO algorithm-L-REVO to realize the front-end real-time operation through lightweight improvement; fusing L-REVO with the back-end of CCMSLAM system to propose a complete multi-machine collaborative SLAM architecture; adjusting the front-end sensors and algorithms to verify the impact on system performance when the front-end is homogeneous or heterogeneous, respectively. On the public dataset TUM, the system improves the localization accuracy by 59. 4% and 31. 6% in both modes, respectively, and the energy efficiency ratio by 8 times compared with the CCMSLAM system. Finally, the system is used for indoor scenario experiments with a front-end power consumption of only 1. 43 W, which verifies the feasibility and effectiveness of the proposed system.

Abstract:To deal with the serious performance degradation of target tracking algorithms in complex tracking scenes, such as background clutter, occlusion, thermal crossover, and target deformation, a real-time infrared single object tracking algorithm based on the adaptive label and sparse-learning correlation filter is proposed. First, sample labels are constructed based on the target response adaptively, and the discrimination ability of the correlation filter is enhanced by training with adaptive sample labels, which suppresses the pollution of the interference region to the tracking model. Secondly, the sparse learning strategy is introduced to suppress the multi-peak distribution of the response map in complex tracking scenes by its L1 norm, resulting in improving the robustness of the tracking algorithm. Compared with the baseline algorithm, the precision and AUC of the proposed algorithm are improved by 19. 3% and 39. 8% , respectively. Experimental results on datasets GTOT, RGBT234, and VOT-2016TIR show that the proposed algorithm has a strong ability to deal with the above complex tracking scenes. Its running speed is over 35 fps, and its comprehensive performance is better than the compared tracking algorithms.

Abstract:As the number of navigating individuals increases in the robot network, how to determine the navigation platform participating in the network and obtain the state and measurement propagation for probabilistic data association are the key to realize small-scale multi-motion platform cooperative navigation. In this article, a multi-platform collaborative navigation method in the network based on the belief propagation for probabilistic data association is proposed. The article uses random finite sets to model states and measurements and formulate the labeled multi-Bernoulli particle filters. A multi-platform collaborative navigation method in the network using the absolute observation of the moving target platform and the sensor station, and the relative observation of the mobile target platform and the adjacent platform, two kinds of observations are data fusion, and the message passing in the system are analyzed by belief propagation on particle filter. The non-parametric belief propagation algorithm extends factor as an approximation of the belief algorithm. Based on non-parametric belief propagation, the belief propagation probability data association on particle filter algorithm is realized to estimate the state of the navigation system. The simulation results show that the non-parametric belief propagation algorithm has poor performance in different base stations and different particle numbers. Comparatively, the proposed algorithm is less affected by the number of base stations and particles and has good robustness and convergence. Compared with the previous method, the algorithm achieves a root mean square error of less than 0. 3 cm, and 10-time increase in accuracy. The location information of the navigation platform can be effectively retrieved.

Abstract:In practical applications, the boiler wall-climbing robot has some problems, such as limited use of wheel encoders and inertial measurement units (IMU), and easy slippage between the wheels and the metal water wall. In the view of the above factors leading to the decline of robot global localization and pose tracking performance, a robot global localization method based on the laser odometry and the improved adaptive Monte Carlo localization (AMCL) is proposed. Firstly, the laser odometry based on PL-ICP method is adopted to replace the traditional wheel and inertia odometry. Secondly, the idea of DNA cross mutation in genetics is introduced into the particle sampling process of AMCL algorithm to design an improved AMCL method based on genetic algorithm, which is able to alleviate the problems of posture tracking performance degradation and slow location recovery caused by AMCL particle depletion. Experimental results show that the absolute localization error of this method is controlled within 12. 7 cm and the accuracy is 32. 4% higher than that of AMCL method. The localization result of this method is almost unaffected when the robot slips slightly. The speed of the system to restore the localization is 35% higher than that of the ordinary AMCL method when the robot slips greatly.

Abstract:When the current source PWM rectifier is at the operate condition that the grid voltage is unbalanced, the negative sequence component of grid voltage will cause low-frequency pulsation in DC side and low-order harmonics in grid side current. To solve this problem, the resonance control with the inner DC current loop based on the traditional double closed loop with active damping is applied in this article to directly suppress low-frequency pulsation existed in the DC side outputs, and the sinusoidal current at the grid side is formed. Besides, the current of the filter capacitor as the compensation term is added into the current reference value on the grid side. In this case, current and voltage with the same phase are realized on the grid side. Furthermore, to improve the operating performance of the system under extremely unbalance conditions, a notch filter is introduced in the modulation link to eliminate low-order harmonic components, which improves the current waveform on grid side. Finally, simulation test is studied by using MATLAB/ Simulink and a 3 kW prototype is established. The simulation and experimental results indicate that when the unbalance degree of the grid reach 6. 7% and 20% , the proposed control strategy can effectively suppress the voltage fluctuation on DC side within 1. 2 V, and the total harmonic distortion rate on the grid side current is less than 4% .

Abstract:Negative matrix factorization (NMF), a tool for matrix factorization and nonlinear dimension reduction, has been widely used for factorization coding and feature extraction of vibration spectrograms from multiple simples. However, the systematic exploration and research on the NMF coding of single vibration spectrogram and the relationships between NMF factorization vector and vibration spectrogram components is still lacking. The basic principles of feature extraction using single spectrogram coding and demodulation are explained. The ability of the NMF to part-based characterize single spectrogram, the bandpass filtering amplitude-frequency characteristics (BFAC) of single spectrogram NMF basis vector, and the synchronous change characteristic between NMF coding vector and spectrogram component are mainly discussed. Two novel feature extraction methods using nonnegative matrix coding and demodulation of single spectrogram are developed, filtering demodulation based on NMF basis vector and demodulation directly performed on NMF coding vector. A BFAC index is defined for adaptive selection of NMF low dimensional parameter, likewise an optimal iteration rule with NMF basis vector normalization for optimization process of NMF coding. The proposed method is applied to analyze the simulated signal and gearbox vibration signal. Under the conditions of given factorization rank and NMF maximum iterations of 300, it takes about 3. 5 s to extract the features of data with the length of 6. 4 k points. Meanwhile, the fault characteristics of the simulated signal with signal-to-noise ratio of -10 dB and that of the gearbox vibration signal with compound faults are extracted.

Abstract:To effectively evaluate the radiation performance of RF front-end circuitries as demanded by the wireless industry, this article proposes a deep learning-based over-the-air (OTA) measurement system. By training a fully-connected deep neural network (FCDNN) with radiation measurements in some test points, we are able to accurately estimate the radiation performance of a RF circuitry in all 3D directions. To balance between the number of radiation measurements for FCDNN training and the estimation accuracy, we further propose to dynamically evaluate the accuracy of the trained model and increase the number of training radiation measurements, until the trained mode can satisfy a predefined accuracy. Experimental results show that the proposed OTA measurement system can accurately reconstruct the radiation performance of a RF circuitry with approximately 60% test points as compared to traditional methods. The proposed OTA measurement system can provide an accurate but cost-effective radiation measurement solution for the wireless industry.

Abstract:As an essential working parameter in crude oil exploitation, the dynamic liquid level depth of oil well is also the indispensable basis for the reasonable arrangement of oil exploitation. In this article, the optimization method based on noise excitation is proposed to ameliorate the existing acoustic resonance method, which has insufficient measurement range and low measurement accuracy. Firstly, the effect of various excitation frequency bands on the measurement results is investigated, and the mathematical model of the output response signal of the system with strong noise excitation is formulated. Then, the extraction algorithm of resonance signal is proposed, which is based on the established mathematical model. The power spectrum estimation and the adaptive homomorphic filtering are applied to suppress the interference of strong noise. Finally, the discrete spectrum correction algorithm based on bilinear interpolation is studied. The resonance characteristic parameters can be estimated accurately. Experimental results show that the resonance characteristic can be extracted under strong noise. The stable measurement of dynamic liquid level can be achieved, where the length of liquid level was over 1 700 m and fluctuation was less than 2 m.

Abstract:In this article, an energy conversion system suitable for the magnesium seawater batteries is designed. To address the problems of soft output characteristics and low voltage of magnesium seawater batteries, the conventional non-isolated three-port converter is optimized to improve the voltage gain. Based on this converter topology. Considering the output characteristics of magnesium seawater batteries and the load characteristics, an energy management strategy for DC/ DC converter system suitable for sudden load change is proposed. By controlling the flexible flow of energy between any two ports in the system, the dynamic power balance of source and load is achieved. Meanwhile, according to the output characteristics of magnesium seawater batteries, a maximum power point tracking method based on the improved grey wolf optimization is introduced into the system energy management, which further improves the energy conversion efficiency of the magnesium seawater batteries. Finally, simulations and experiments demonstrate that when the load power changes between 0. 36 and 1. 2 W, the converter can complete the mode switching accordingly. On the premise of ensuring the output voltage is stable at 12 V, the flexible control of the output current of 30 and 100 mA is realized, which verifies the rationality of the hardware and software design of the energy conversion system.

Abstract:With the growth of demand and the improvement of process complexity in nuclear power, petrochemical and other industries, the safety monitoring of the status of key equipment lacks direct, real-time and online monitoring methods. It is difficult to achieve accurate diagnosis and early warning of faults. In this article, the shell and tube heat exchanger, key equipment in the petrochemical and nuclear power industries, is selected as the research object. By analyzing the fouling mechanism of the heat exchanger and its impact on the heat transfer performance and applying the total heat transfer coefficient to establish the characterization relationship among the heat transter efficiency, the heat transfer mean temperature differential and the fouling thickness, a virtual sensing method of the fouling of the shell and tube heat exchanger is proposed. The method uses high-precision and easily accessible operation data of the heat exchanger, including temperature, pressure and flow, to realize the online accurate sensing of the fouling thickness of the heat exchanger. Finally, test is implemented by using a high-pressure vessel test piece on the hot state functional test bench circuit. The results show that the virtual sensing model of scale in heat exchanger based on the total heat transfer coefficient can monitor the fouling thickness above 2. 2% of the tube diameter within a 30% error range.