Office Online  
    Office Online
    Manuscript Tracking
    Editor Work
    Office Work
    Editor-in-chief
Journal Online
    Current Issue
    Advanced Search
    Archive
    Read Articles
    Download Articles
    Email Alert
Traffic
Quick Search  
  Adv Search
2021 Vol. 42, No. 7
Published: 2021-07-28
 
833 Fast Positioning Algorithm for End-effector of Cooperative Manipulator
YU Jing,JIANG Wen-song,LUO Zai,YANG Li,ZHOU Gen-ming
DOI: 10.3969/j.issn.1000-1158.2021.07.01
Aiming at the poor real-time positioning effect of the traditional cooperative manipulator end-effector, AprilTag tag positioning was adopted to propose a fast positioning algorithm based on geometry to extract ROI. The algorithm uses a downsampling method to search the effective local range for macro-size targets, and establishes geometric constraints to extract edge pixels to accelerate the speed of target recognition. In order to verify the effectiveness of the proposed algorithm, a visual positioning system experiment is designed for the end-effector of the manipulator. The experimental results show that the speed of the proposed algorithm is improved to 40frame/s under the average relative error of corner detection is lower than 0.1pixel. Therefore, the proposeed algorithm realizes the rapid positioning of the end of the manipulator.
2021 Vol. 42 (7): 833-839 [Abstract] ( 297 ) HTML (1 KB)  PDF (3825 KB)  ( 522 )
840 A Visual Model Based on Attention Mechanism and Convolutional Neural Network
LI He-xi,LI Ji-hua,LI Wei-long
DOI: 10.3969/j.issn.1000-1158.2021.07.02
To solve the problems of current deep convolutional neural network (CNN), such as large model size, many training parameters, slow computing speed, and difficulty in transplantation to mobile terminal, a visual model of depthwise separable convolution with triple attention module (DSC-TAM) is proposed. Firstly, the depthwise separable convolution is used to reduce the model parameters and improve the computing speed of the network model. Secondly, the triple attention mechanism module is introduced to improve the ability of feature extraction and network performance. The experimental results show that the recognition rate of this method is 99.63%, the model size is reduced by 13%. Compared with the standard convolutional neural network visual model and other methods, the recognition accuracy is guaranteed, and the size of the network model is reduced.
2021 Vol. 42 (7): 840-845 [Abstract] ( 254 ) HTML (1 KB)  PDF (1813 KB)  ( 508 )
846 Obstacle Detection Method Based on Vehicle 16-line Lidar
KONG De-ming,DUAN Cheng-xin,GOOSSENS Bart,WANG Shu-tao
DOI: 10.3969/j.issn.1000-1158.2021.07.03
Aiming at the issue of low accuracy of the existing in obstacle detection algorithm in the vehicle 16 line Lidar point cloud data, an obstacle detection algorithm based on adaptive grid is proposed. Firstly, octree and random sample consensus (RANSAC) algorithm is utilized to remove the ground point.Secondly, project the point cloud onto the 2D-grid, tall structure objects can be quickly extracted based on the elevation information in each grid.Thirdly, a two-level grid model is established, the sub-grid resolution is determined adaptively according to the distribution information of coarse grid clustering results, the obstacles that may contain multiple targets are detected precisely at the sub-grid layer.Finally, the clustering results are improved by combining the state information of two adjacent obstacles.The experimental results under urban road environment test sets show that the proposed method can precisely detect obstacles in driving area, the optimized clustering algorithm can reduce the error rates of under-segmentation and over-segmentation,the detection accuracy is 91%.
2021 Vol. 42 (7): 846-852 [Abstract] ( 332 ) HTML (1 KB)  PDF (2872 KB)  ( 661 )
853 Prediction of Steam Turbine Heat Consumption Rate Based on Improvement Lion Swarm Optimization
WANG Chan-chan
DOI: 10.3969/j.issn.1000-1158.2021.07.04
Aiming at the problem that the model of steam turbine heat consumption rate is difficult to be predicted accurately, a method based on the improved lion group algorithm and fast learning network integrated modeling was proposed. Firstly, the traditional lion swarm optimization algorithm is prone to premature convergence and the slow speed of the algorithm in the late iteration leads to the algorithm falling into the local optimal defect. The algorithm was improved by introducing the tabu search, the nonlinear disturbance factor and the golden sine strategy. Secondly, the improved lion swarm algorithm was numerically validated, and the results showed that it had the higher convergence accuracy and the faster convergence speed. Finally, a prediction model of steam turbine heat consumption rate was established based on the operation data of steam turbine in a thermal power plant, and a fast learning network optimized by improved lion swarm optimization algorithm was used to predict the heat consumption rate. The experimental results were compared with other optimization strategies, and the results showed that the fast learning network prediction model based on improved lion swarm algorithm had the higher generalization ability and improved the prediction accuracy of steam turbine heat consumption rate.
2021 Vol. 42 (7): 853-860 [Abstract] ( 292 ) HTML (1 KB)  PDF (1526 KB)  ( 445 )
861 Identification of Two-phase Flow Pattern Based on the Sparsity of Measured Capacitance for Electrical Capacitance Tomography
ZHANG Li-feng,MIAO Yu
DOI: 10.3969/j.issn.1000-1158.2021.07.05
The flow pattern identification algorithm based on the sparsity of measured capacitance for electrical capacitance tomography (ECT) is proposed. Firstly, the over-complete dictionary of normalized capacitance measurement for investigated flow patterns is built, with which the sparse representation of the sample can be obtained. And then, the basic requirements of sparse reconstruction can be met. The orthogonal matching pursuit (OMP) algorithm is used to calculate the sparse solution of each standard sample using the over-complete dictionary.Finally, the flow pattern is identified according to the linear correlation between the sample to be identified and the sparse solution of the standard sample. Simulation and static experiments are carried out for the five typical two-phase flow patterns, and the correct identification rate is higher than 98%.
2021 Vol. 42 (7): 861-865 [Abstract] ( 212 ) HTML (1 KB)  PDF (930 KB)  ( 174 )
866 The Simulation of Atmospheric Detached Shock Wave of Supersonic Multi-hole Porous Probe
ZHANG Yang-chun,ZHOU Shu-dao,YAO Tao
DOI: 10.3969/j.issn.1000-1158.2021.07.06
In order to study the change law of the shape and standoff distance of the shock wave for hemispherical porous probe under supersonic atmospheric conditions, when the Mach number changes. CFD method was used to simulate the atmospheric environment, and numerical simulation experiments were carried out in the range of 1.2～1.7 Mach number. Hyperbola was used to represent the detached shock wave, and the functional relationship between the curvature, the standoff distance, the directrix distance of the shock wave and the Mach number was fitted by the least square method, and the parametric equation of the detached shock wave curve was established. The parametric equation is compared with the simulation results and the empirical formula, and the results show that the parametric equation is in good agreement with the simulation results, and the accuracy is higher when the distance is closer to the sphere. The difference between the distance predicted by the parametric equation and the empirical formula does not exceed 5% of the sphere radius.
2021 Vol. 42 (7): 866-872 [Abstract] ( 262 ) HTML (1 KB)  PDF (5042 KB)  ( 175 )
873 A Data Fusion Method of Double-channel Ultrasonic Flowmeter Application in low pressure Gas
ZHAO Wei-guo,BU Qin-chao,YAO Hai-bin,ZHANG Sheng-yi,ZHANG Tao
DOI: 10.3969/j.issn.1000-1158.2021.07.07
The double-channel ultrasonic flow measurement is conventionally applied in low-pressure gas of the small and medium pipes. However, the error data and wrong data of one channel will be produced by the high attenuation and low signal-to-noise ratio when the ultrasonic signals travel in the low-pressure gas. So the accuracy and stability of ultrasonic flowmeter will be decreased. In order to solve these problems, a fusion method of double-channel ultrasonic flowmeter based on time-difference is proposed. Firstly, the gross errors of time-difference data are eliminated and flowrate is calculated in every channel. Then, the state of the flowrate is estimated by time-difference data. Finally, the improved Kalman data fusion is adopted to calculate the mean flowrate in pipe. The experimental results show that the measurement error is -0.58% and the repeatability is 0.21%.
2021 Vol. 42 (7): 873-878 [Abstract] ( 228 ) HTML (1 KB)  PDF (1679 KB)  ( 551 )
879 Image Reconstruction Algorithms of ECT System Based on the Modified Linearized Bregman Algorithm
MA Min,SUN Mei-juan
DOI: 10.3969/j.issn.1000-1158.2021.07.08
For the underdetermination of the inverse problem of electrical capacitance tomography (ECT), the improved linear Bregman algorithms based on singular value decomposition (SVD) and second-order iteration were applied to the ECT image reconstruction process. The simulation experiments showed that compared with SVD, Landweber, CG and linear Bregman algorithm, they had the higher image resolution and faster imaging speed. Among the two improved algorithms, the algorithm based on singular value decomposition had fewer image reconstruction artifacts, and the algorithm based on second-order iterations significantly increased the speed.
2021 Vol. 42 (7): 879-884 [Abstract] ( 222 ) HTML (1 KB)  PDF (1065 KB)  ( 142 )
885 Finite Element Analysis and Experimental Analysis for 200kW/180MJ Flywheel Energy Storage System
JIN Mei,YAN Ting-xin,ZHANG Li-guo,LI Guang-jun,WANG Wei,WANG Na
DOI: 10.3969/j.issn.1000-1158.2021.07.09
A 200kW/180MJ flywheel energy storage system is designed to solve the problem of large load fluctuation of the power grid system. The finite element method is used to model and simulate the flywheel shafting, and the critical speed and vibration mode of the flywheel rotor are calculated, and the modal and harmonic responses of the flywheel rotor supported by mechanical bearings are analyzed. The full cycle detection of the flywheel speed rise and fall is carried out, the amplitude-frequency characteristics of the flywheel running process are recorded, and the amplitude-frequency characteristics of the flywheel are compared and analyzed. The flywheel energy storage system is running well without friction. The oscillation amplitude’s maximum analysis result is 1.6mm less than the designed gap of 3mm.
2021 Vol. 42 (7): 885-891 [Abstract] ( 249 ) HTML (1 KB)  PDF (4092 KB)  ( 552 )
892 Bearing Fault Diagnosis Method Based on VMD and Convolutional Neural Network Undervarying Operation Conditions
CHEN Jian,HUANG Kai-xuan,Lü Wu-yang,LIU Yuan-yuan,YANG Bin,LIU Xing-fu,CAI Kun-qi
DOI: 10.3969/j.issn.1000-1158.2021.07.10
To investigate the problem that it was difficult to obtain a large number of bearing fault data and diagnosefault type under varying operation conditions, a bearing fault diagnosis method based on variational mode decomposition and convolution neural network was proposed. This method could diagnose bearing data under varying operation conditions by using training data under steady conditions.Firstly, variational mode decomposition was used to decompose the bearing vibration signals in order to obtain a series of band-limited intrinsic modal functions.Then, convolution neural network was constructed to adaptiveextract and classifiy featuresof the IMFs, with optimization technology used to improve its adaptability.Finally, the rolling bearing fault data obtained from bench test was used in experimental verification, and the model of ResNet and SVM were used as comparison. The results showed that the diagnosis/recognition rate of the model is 100% / 98.86%under varying operation conditions that is higher than two comparison models, which also proved that the model can effectively realize bearing fault diagnosisunder varying operation conditions.
2021 Vol. 42 (7): 892-897 [Abstract] ( 250 ) HTML (1 KB)  PDF (2531 KB)  ( 808 )
898 Bearing Fault Diagnosis Based on VMD Energy Entropy and Optimized Support Vector Machine
JIN Jiang-tao,XU Zi-fei,LI Chun,MIAO Wei-pao,LI Gen
DOI: 10.3969/j.issn.1000-1158.2021.07.11
The early fault signals of rolling bearings are relatively weak, and are affected by the coupling of noise and vibration, which leads to inaccurate fault judgments. Based on variational mode decomposition (VMD) and energy entropy, multi-mode characteristic matrix is constructed. Grey wolf optimizer (GWO) is adopted to optimize the parameters of support vector machine (SVM). VMD-Entropy-OSVM bearing intelligent fault diagnosis is proposed, using bearing experimental data to verify the effectiveness and superiority of the proposed method. The experimental results show that VMD-Entropy-OSVM not only recognizes different fault types at the end of bearing damage, but also has high accuracy at the beginning of bearing damage. The accuracy of the proposed method is up to 99.8% at 8dB, which is 3.3%～27.3% higher than the existing method. When the SNR is 0dB, the accuracy is still 73.5%, which is 11%～33% higher than the existing method, the model shows good generalization performance. In addition, the running time is shorter and more efficient under the same computing resources.
2021 Vol. 42 (7): 898-905 [Abstract] ( 336 ) HTML (1 KB)  PDF (3647 KB)  ( 397 )
906 Research on Cold Rolling Force Prediction Model Based on Improved Deep Belief Network
WEI Li-xin,WANG Heng,SUN Hao,HU Zi-yu
DOI: 10.3969/j.issn.1000-1158.2021.07.12
In the cold rolling process of strip steel, the accuracy of rolling force prediction directly determines the rolling precision and product quality of the strip. The traditional single-hidden layer-based neural network modeling method is simple in structure, and the expression ability and generalization ability of complex functions are restricted. The rolling site environment is complex, and data measurement has noise interference,which will directly affect the forecasting accuracy. Regarding the issue above, an improved deep belief network prediction model based on unsupervised learning is proposed. The construction of denoising-restricted Boltzmann machines and deep networks can improve the systems ability to learn the characteristics of input data, while training the deep network with improved contrast divergence algorithm.Finally,the model is tested by using the measured data of a steel mills 1200mm rolling mill, and three different models are compared and analyzed. The results show that the average relative error of the rolling force prediction of the model is controlled within 4.5%, and the time required for modeling is reduced by 26% compared to the self-encoding network.
2021 Vol. 42 (7): 906-912 [Abstract] ( 206 ) HTML (1 KB)  PDF (1817 KB)  ( 519 )
913 Periodicity and Symmetry in Quantization Error of Phase Angle of Sinusoidal Waveform Generated by Digital-to-Analog Converter
LU Zu-liang
DOI: 10.3969/j.issn.1000-1158.2021.07.13
A sinusoidal waveform generated by digital-to-analog converter is actually a fundamental of staircase waveform. The quantization error of phase angle of this kind of the sinusoidal waveform is defined. Its intrinsic characteristics such as periodicity and symmetry are described and proved. The relative zero-point distribution is derived. The application of these characteristics in building of phase angle with higher accuracy is discussed. This zero-point distribution is quantization too with a width of 0.5H (H=2π/N) or 0.25H for odd numbers of N. In order to overcome the quantization error of phase angle, it is advised to the phase angler become an integer multiple, p, of this width, where parameters of p and N can be adjusted to such that an arbitrarily phase angle difference demand can be realized in principle within a specified uncertainty.
2021 Vol. 42 (7): 913-922 [Abstract] ( 281 ) HTML (1 KB)  PDF (1766 KB)  ( 185 )
923 Equivalent Modeling of Micro-grid using Optimized ESN
WU Zhong-qiang,QI Song-qi,SHANG Meng-yao,SHEN Dan-dan
DOI: 10.3969/j.issn.1000-1158.2021.07.14
In order to make micro-grid modeling more accurate, an equivalent modeling method of micro-grid based on echo state network (ESN) is proposed. Under various operating conditions of the micro-grid, the micro-grid equivalent model based on the echo status network is constructed with the current and power data of the access terminal, which are taken as the input and output of the network respectively. Since the initialization parameter of echo state network no longer changes, and lacking adaptability, it lead to the inability to achieve optimal approximation. Fireworks algorithm has the advantages of explosiveness, instantaneousity, parallelism and scalability. In order to improve the accuracy of the modeling, the fireworks algorithm is used to optimize the parameters of the echo state network, a mathematical model by simulating the explosion of fireworks is established, and selects the best individual by calculating individual fitness values. By comparing with the measured simulation data of the micro-grid connected the grid, the rationality and accuracy of the modeling method are verified, which shows that the model has a good practical value.
2021 Vol. 42 (7): 923-929 [Abstract] ( 192 ) HTML (1 KB)  PDF (2614 KB)  ( 195 )
930 A Novel Digitizing Evaluation and Traceability Method of Modulation Meters
LIANG Zhi-guo,HE Zhao,MIAO Jing-yuan,GUO Xiao-tao,ZHANG Yi-chi
DOI: 10.3969/j.issn.1000-1158.2021.07.15
For the modulation measuring instrument, a calibration method is proposed, which includes directly using a digital oscilloscope for signal acquisition and measurement in the case of low-frequency carrier, and performing software demodulation analysis to obtain the modulation domain parameters. For the high-frequency carrier case, it is first converted to a low-frequency carrier case using down-conversion frequency technology, and then measured and analyzed. All the courses are discussed in details. In this way, the measurement value of the modulation domain instrument can be traced to the sampling rate and amplitude of the digital oscilloscopes, so as to achieve the purpose of calibrating the modulation meters. Calibration verification experiments performed on a specific type of instrument validate the effectiveness and feasibility of the method described herein.
2021 Vol. 42 (7): 930-936 [Abstract] ( 225 ) HTML (1 KB)  PDF (1816 KB)  ( 228 )
937 Batch Hierarchical Coding Extreme Learning Machine Based on Manifold Regularization
LIU Bin,YANG You-heng,LIU Jing,WANG Wei-tao,LIU Hao-ran,WEN Yan
DOI: 10.3969/j.issn.1000-1158.2021.07.16
Aiming at the problems of high memory energy consumption,low classification accuracy and poor generalization when dealing with high-dimensional data, a batch hierarchical coding extreme learning machine algorithm is proposed.Firstly,the dataset is processed in batches to reduce the data dimension and reduce input complexity.Then,the multi-layer automatic encoder structure is used to unsupervise the batch data to achieve deep feature extraction.Finally,the manifold regularization is used to construct a manifold classifier with inheritance factors to maintain data integrity and improve the generalization performance of the algorithm.The experimental results show that the method is simple to implement,and the classification accuracy on the NORB,MNIST,and USPS datasets can reach 92.16%,99.35%,and 98.86%,respectively.Compared with other ELM algorithms,it has obvious advantages in reducing computational complexity and reducing CPU memory consumption.
2021 Vol. 42 (7): 937-943 [Abstract] ( 213 ) HTML (1 KB)  PDF (735 KB)  ( 174 )
944 Selection and Application of Calibration Parameter Optimization for Traction Measurement System with Large Reflective Load
ZHANG Li-fei,DU Jing,WANG Yi-bang,LUAN Peng,WU Ai-hua,LIANG Fa-guo
DOI: 10.3969/j.issn.1000-1158.2021.07.17
Compared with traditional load-pull measurement systems, the active hybrid load-pull measurement system can achieve a larger load impedance adjustment range for microwave power amplifier device testing, which has become a popular measurement solution.For the verification of active hybrid load-pull measurement system, the transducer gain is generally selected as the optimization verification parameter of the load traction measurement system.By comparing the deviation between the theoretical value and the measured value of the power gain and the transducer gain, it is observed that under large reflection conditions, the power gain is more sensitive for the residual error after system calibration, and the effect of optimizing and improving the system accuracy is better.Finally, it is verified by the application experiment of power optimization, it is observed that when the load reflection coefficient reaches 0.9, the transducer gain changes 0.4dB, whereas the power gain changes 0.8dB.Therefore, the effect of selecting the power gain as a large reflection optimization calibration parameter is more obvious.However, under a small reflection coefficient, the transducer gain can be selected as the verification parameter of the load traction measurement system.
2021 Vol. 42 (7): 944-949 [Abstract] ( 235 ) HTML (1 KB)  PDF (2613 KB)  ( 347 )
950 Determination of Resin Content in Silicone Glass Fiber Composites by Density Method
Lü Hui,MA Kai-bao,LIANG Xu,CAI Chen,LIU Xin,XUN Qi-ning,GONG Wei,YAO Xu-xia
DOI: 10.3969/j.issn.1000-1158.2021.07.18
Burning method is widely used in determining the resin content of glass fiber composite materials. The burning method can not be used to determine the resin content in the glass fiber composite of organosilicon resin, as the resin base can not be completely removed by burning.Density method is to calculate the resin content by measuring the density of glass fiber, sample and resin. Two different method are used to test the resin content of same batch of glass fiber composite materials. According to the results, the resin content of the density method is 40.8% and the burning method is 38.6%. The expanded uncertainty of the density method is 1.2% and the burning method is 0.3%. The accuracy of density method is slightly lower than that of burning method, but it can be used to determine the resin content of silicone resin glass fiber composites and meet the index requirements. Therefore, the density method can replace the burning method to test the samples, but the experimental process needs to be strictly controlled to improve the accuracy of the test results.
2021 Vol. 42 (7): 950-955 [Abstract] ( 253 ) HTML (1 KB)  PDF (408 KB)  ( 297 )
956 Reliability Evaluation of Gas Dynamic Dilution Device
WU Yu-tong,BI Zhe,YANG Yang-zhongfu,MA Hao-miao,WU Hai
DOI: 10.3969/j.issn.1000-1158.2021.07.19
A dilution device was built based on mass flow controllers (MFC), and a high-precision CH4 spectrometer was used to evaluate the reliability of the gas dilution device. Experimental results showed that when the dilution ratio was 0.01～0.02, the relative error of the theoretical concentration of gas mixture produced by the device was not more than ±4%, and when the dilution ratio was 0.04～0.1, the relative error was within±0.5%. The accuracy of the dilution device was not so good when the dilution ratio was very small. The change of input gas pressure upstream of MFC had influence on its output gas flow. When the input pressure changed in the range of 0.2 ~ 0.6MPa, the variation range of MFC output gas flow was not more than 1%, resulting in the variation of standard gas concentration output by the dilution device was not more than 1%. When the input gas pressure changed in the range of (2～6)bar, the variation range of MFC output gas flow was not more than 1%. The input gas pressure variation could interfere with the diluted gas concentration generated by the dilution device, of which the influence was less than 1%. When using the dilution device to prepare standard gas, it is preferable to set the MFC in the range above 10% of its full scale, so as to ensure the reliability of the dilution device.
2021 Vol. 42 (7): 956-963 [Abstract] ( 307 ) HTML (1 KB)  PDF (594 KB)  ( 226 )
964 Research on Conversion Factors of the Certified Reference Materials on Different Chemiluminescence Immunity Platforms
TANG Yao-xu,YANG Xiao-dong,ZHANG Xuan-pu
DOI: 10.3969/j.issn.1000-1158.2021.07.20
In order to find out the conversion factors between chemiluminescence immunoassay systems of different manufacturers which can be used as reference for indoor quality control and inter-room quality assessment. 19 kinds of national certified reference materials were tested as samples on 181 instruments from 9 manufacturers and the results were analyzed by statistical method. The results demonstrate that tPSA and TSH projects can be compared among different systems with the conversion factors, CEA projects can be compared among a part of different systems with the conversion factors, HCG projects can not be compared among different systems with the conversion factors, HBsAg projects can be compared among different systems with the conversion factors in different concentration ranges respectively, the comparability of AFP projects among different systems needs to be further studied.
2021 Vol. 42 (7): 964-970 [Abstract] ( 236 ) HTML (1 KB)  PDF (409 KB)  ( 320 )
971 Research on the Minimum Room Size Based on D-D Neutron Source Neutron Dose Equivalent Meter Calibration
TIAN Xing-yu,ZHANG Xiong-jie,TANG Bin,HU Bin,SONG Le-tian
DOI: 10.3969/j.issn.1000-1158.2021.07.21
In the calibration process of neutron peripheral dose equivalent instrument, using D-D neutron source instead of traditional isotope neutron source will make the calibration process more secure. In order to promote the application of D-D neutron source in the calibration process of neutron peripheral dose equivalent instrument, the method of scattering neutron research using MC energy truncation instead of shadow cone method is proposes, and the influence of different internal space sizes on scattering neutron in different types of rooms is analyzed. The calculation results show that with the increase of the internal space of the room, the scattering neutron of the incident detector accounts for the proportion will gradually decrease. If the D-D neutron source is used for the calibration of the neutron peripheral dose equivalent instrument, when the detection distance is 75cm and the detector diameter is 20cm, the minimum cube room required is a cube room with an internal space of 332cm edge length; the minimum cube room required is a cuboid room with an internal space of 410cm in length, 410cm in width and 205cm in height.
2021 Vol. 42 (7): 971-976 [Abstract] ( 227 ) HTML (1 KB)  PDF (1136 KB)  ( 155 )
News
Download
Links
Copyright © Editorial Board of Acta Metrologica Sinica
Supported by:Beijing Magtech