Research On In-situ Test Technology For Electromagnetic Characteristics Of Electrical Equipment And System

The in-situ test of electromagnetic characteristics of electrical equipment and system is an engineering problem to be solved.It is of great significance to the EMC prediction test of electrical equipment and system and the elimination of electromagnetic fault,and needs to be further studied.On the basis of fully understanding the development status and existing problems of electromagnetic characteristics in-situ test of electrical equipment and system,this paper mainly carries out the research on electromagnetic characteristics in-situ test technology of electrical equipment and system,aiming to suppress the electromagnetic background noise of electromagnetic characteristics in-situ test of electrical equipment and system,and realize the synchronous separation of electromagnetic characteristics of multiple equipment under positive and underdetermined conditions,And build the electromagnetic characteristics of electrical equipment and system on-site testing platform to complete the electromagnetic characteristics of electrical equipment and system on-site testing tasks under different conditions.The specific research contents include the following aspects:Firstly,an improved intelligent adaptive filtering algorithm is proposed to solve the problem of electromagnetic characteristics in-situ test of single equipment under background noise.A new multi-layer adaptive filter structure is designed.After the broadband electromagnetic signal is decomposed by multi-scale full band,the improved adaptive filtering algorithm based on particle swarm optimization is processed,It solves the problem of the imbalance of convergence speed and steadystate error of traditional algorithm,improves the processing ability of broadband electromagnetic radiation signal,and realizes the elimination of background noise in the in-situ test of electromagnetic characteristics of single equipment.The in-situ test results are compared with the anechoic chamber test results to verify the effectiveness of the algorithm.Secondly,aiming at the problem of synchronous testing of electromagnetic characteristics of multi equipment under the condition of underdetermined electrical equipment and system,this paper proposes a source number estimation method based on support vector machine and Gaishi circle,which improves the success rate of in-situ testing.A comprehensive filtering algorithm based on kurtosis and correlation is proposed to filter the background noise and other incoherent components,and compress the data dimension on the basis of preserving the original signal information.The structure of multi-channel underdetermined blind electromagnetic radiation synchronous separation algorithm is designed,which breaks through the limitation of the number of signal channels,saves the test space and construction cost,improves the efficiency of multi equipment electromagnetic characteristics in-situ test,and realizes the fast and accurate separation of multi-target source signals under underdetermined conditions.The simulation and real ship test data verify the effectiveness of the algorithm for synchronous separation of electromagnetic characteristics of multi equipment under the condition of underdetermined.Furthermore,to solve the problems of slow convergence and poor steady-state performance of the positive definite blind source separation method,this paper proposes a positive definite blind source separation method based on maximum likelihood estimation criterion and neural network.The improved objective function of blind source separation and neural network optimization algorithm with double acceleration strategy are adopted,The neural network loss function decreases rapidly along the flat path,improves the convergence speed of the positive definite blind separation algorithm,reduces the steady-state error of the algorithm,and realizes the synchronous separation of electromagnetic characteristics of multiple devices under the positive definite condition.Simulation and experimental data verify the superiority of the algorithm.Finally,this paper carries out the research on the implementation of the in-situ test platform for the electromagnetic characteristics of electrical equipment and system.This paper proposes an implementation technology of the in-situ test system based on the integration of software and hardware.It uses parallel distributed computing and recursive online processing technology to optimize the in-situ test algorithm,improves the execution speed of the in-situ test platform,and designs the operation mode of multi-mode switching,The application scope of electromagnetic characteristics in-situ test platform for electrical equipment and system is expanded.The electromagnetic characteristics test platform of electrical equipment and system is designed to meet the electromagnetic characteristics test requirements of electrical equipment and system under different conditions.The results of in-situ test under different conditions verify the correctness of in-situ test platform for electromagnetic characteristics of electrical equipment and system.The research results of this paper have important theoretical research significance and engineering application value,and are of great significance for the in-situ test research of electromagnetic characteristics of electrical equipment and system,the suppression of electromagnetic interference and the design of in-situ test platform.The research results can be widely used in the electromagnetic characteristics test of electrical equipment and systems such as national defense weapons,aerospace,ships and so on.