Research On Measurement Method Of Low Frequency Magnetic Field Radiation In Maglev System

As a new type of transportation,high-speed maglev has not yet been popularized in the world.However,as a means of transportation involving system integration,vehicles,traction power supply,operation control communication,suspension traction,etc.,its international strategic significance is of great significance with a high degree of engineering complexity.The increasingly complex electromagnetic environment and the characteristics of maglev trains pose a huge threat to the electromagnetic compatibility of it.Aiming at the problems existing in the measurement method of the low-frequency magnetic field radiation signal of the high-speed maglev train,this dissertation studies it based on the blind source separation model.This dissertation firstly summarized the research status of magnetic field radiation and blind source separation in high-speed maglev trains in detail,and determined the research route of this dissertation for the problems existing in blind source separation in low-frequency magnetic field radiation measurement of high-speed maglev trains.Aiming at the amplitude uncertainty existing in the blind source separation algorithm,an amplitude estimation method based on the least squares method was proposed.For the problem that the number of source signals is unknown,this dissertation adopted the method of directly calculating the correlation coefficient matrix to identify the estimated signal.So,the complicated calculation process was avoided and it was verified experimentally.For the main emission source of high-speed maglev train-long stator synchronous linear motor,a Maxwell3 D three-dimensional model was established,and the magnetic field radiation was simulated and analyzed.The current relevant standards were analyzed in detail to provide a certain reference for the electromagnetic compatibility standards of high-speed maglev trains.Combined with the field measured data to evaluate its radiation limit,the results showed that the low-frequency radiation disturbance of the maglev train meeted the current standards.In view of the fact that the measured data is easily disturbed by background noise,wavelet noise reduction is introduced in this dissertation to preprocess the data,and the effectiveness of the blind source separation algorithm based on wavelet transform(WT-Fast ICA)was verified by designing a control experiment.In order to improve the accuracy of the separated signal,Constrained independent component analysis(c ICA)was introduced to analyze the low-frequency magnetic field radiation signal of the maglev train,and the wavelet decomposition was used to construct a reference signal,which was applied to the separation of frequency domain signals.Experiments showed that the constrained independent component analysis(WD-c ICA)algorithm based on wavelet decomposition had a significant effect on the extraction of magnetic field radiation spectrum signals.Finally,the blind source separation software was designed based on MATLAB,and the effectiveness of the software was verified by the detection of simulated and measured signals.