Font Size: a A A

Design And Research Of Satelite Power System Fault Diagnosis Platform

Posted on:2019-10-03Degree:MasterType:Thesis
Country:ChinaCandidate:Y ZhangFull Text:PDF
GTID:2382330566497767Subject:Instrumental Science and Technology
Abstract/Summary:PDF Full Text Request
With the wide application of satellites and the diversification of their functions,fault diagnosis technology of satellite power system which is the heart of satellite,has become a hot topic in relevant fields both at home and abroad.The fault diagnosis of satellite power system is mainly divided into two aspects: fault detection and fault identification,and the research of related technologies is based on a large number of data.However,the satellite power system has less telemetry data and poor real-time transmission,and the number of fault samples that the researchers can obtain is particularly limited,which has brought unfavorable factors to the research in this field.Therefore,this paper studies from three aspects of system modeling and simulation,optimizing diagnosis methods and building diagnostic platform to improve the related research status.In this paper,the power system of geosynchronous satellite was selected as the research object,the working principle of each component was analyzed,a simplified mathematical model was set up,and the simulation model of the satellite power system was built in the Simulink simulation environment.The fault generation mechanism of the satellite power system was analyzed.The typical fault mode was selected to infuse the simulation model,and the simulation data that can reflect the real fault samples can be obtained.The working state of satellite power system was analyzed,and the different periods of its work were divided,and the characteristics of faults in different periods were analyzed.According to the fault characteristics of the satellite power system,the fault detection method based on the principle of wavelet transform and the principle of principal component analysis was adopted.Faults were identified by the probability neural network and the decision tree based on the ID3 algorithm,and the training of neural network and the extraction of the decision tree diagnosis rules were realized.The characteristics of the fault and the dependency of the diagnosis algorithm were described.The fault simulation platform of satellite power system based on MATLAB and its user interface were designed.The function and work flow of the diagnosis platform were introduced.The usage description of the user interface was given,and the reliability of the fault diagnosis platform was tested and analyzed.
Keywords/Search Tags:Satellite power system, Fault diagnosis, Wavelet transform, Principal component analysis, Probabilistic neural network, Decision tree
PDF Full Text Request
Related items