Research On Digital Anechoic Chamber On-site Separating Method For Complex Electromagnetic Environment Of Marine Electric Propulsion System

With the continuous development of society,the degree of modernization and informatization of ships is constantly improving.The electromagnetic radiation in ships is becoming more and more serious,and the electromagnetic environment is getting worse and worse,which has affected the normal operation of ship electric propulsion system and radar communication system.Therefore,new requirements are put forward for EMC testing of marine electric propulsion system.And all kinds of large-scale equipment can not enter the traditional test platform for electromagnetic compatibility testing due to volume reasons.Digital anechoic chamber technology provides an effective way to solve this problem.Signal processing algorithm is the core of digital anechoic chamber.Aiming at the characteristics of multi-channel,non-stationary,overlapping frequency spectrum and strong interference of electromagnetic radiation signals from ship integrated electric propulsion system,a new method of field synchronous testing of digital anechoic chamber based on wavelet analysis(WT)and independent component analysis(ICA)is proposed.In this paper,the wavelet analysis is deeply studied,and the background noise is removed by using the wavelet threshold denoising method.This method optimizes and expands the traditional wavelet threshold denoising,improves and optimizes the threshold function,overcomes the shortcomings of soft and hard threshold methods in reconstructing signals,and has obvious noise suppression effect because of the natural advantages of the wavelet analysis algorithm in principle.It is better than the traditional fast Fourier transform and windowed Fourier transform.It purifies the "outfield" environment and provides a non-electromagnetic interference environment for the blind signal processing.Then the Blind Source Separation(BSS)is used to separate the mixed electromagnetic radiation signals from the equipment.The main method of blind source separation is independent component analysis(ICA).In this paper,FastICA algorithm is used to overcome the limitation of traditional empirical mode decomposition(EMD)method in multi-channel data fusion,which lacks theoretical basis,and the problem of slow data processing and low efficiency.Finally,using MATLAB modeling and simulation,through the Simulink construction system,the purpose of filtering and separating signals is realized.Through the collection and processing of the ship generator data,the electromagnetic radiation spectrum of the equipment is obtained.In this way,electromagnetic compatibility analysis,electromagnetic protection and spectrum management can be performed on the equipment.The research in this paper overcomes the disadvantage of traditional anechoic chamber which is not convenient to test large equipment,improves the efficiency of ship EMC testing,and provides a new method and idea for digital anechoic chamber in the future.