The Research Of Satellite Fault Diagnosis Based On The Processing Of Signal

The faults detection and diagnosis technique plays an important role in the modernization construction,especially the faults detection in the complex system.The faults are of different kinds and diversified.Utilizing the traditional faults detection method to diagnose and monitor the complex system is booming.It is the emphasis and hot sports of the research to apply the fault diagnosis theory method into the detection and monitoring in the practical system.The development of the theory props up the improvement and innovation of the practical programs.This thesis is on the basis of the"Fault Diagnosis Expert System"(Item Number:Y5W3070301)and the specific diagnosis and health management requested by the engineering center of the micro satellite in Shanghai on the related research on the overall health management of the satellite among the spacecrafts.The spacecrafts like the satellite become increasingly complicated.And in order to reduce the burden of that staff operating the satellite at the greatest extent,and due to the unique features of the micro satellite compared to the traditional satellites,and because of the limited resources on the planet,it is urgent to maintain the highly autonomy and reliability of the diagnosis system and monitoring strategy in terms of satellite.There is another issue yet to be addressed,which was to meet the requirements of this kind of special system diagnosis of the spacecrafts.At present the anomaly detection and repair of the spacecrafts like the satellite is always conducted by manual work on the ground with low efficiency.The operating staffs on the ground adopt all kinds of statistical techniques and models to judge the current operating state of the satellite through receiving and anglicizing the telemeasuring data.Under that abnormal circumstance it is prone to lead erroneous judgement or the misoperation caused by erroneous judgement.Therefore the primary goal is to combine the techniques on the basis of signal processing and the mixed method on the basis of intelligence in order to enhance the efficiency,accuracy and reliability of trend analysis and diagnosis.The following practical design and research focuses on the above points.(1)The preliminary task is to analyze the overall operating state of the system,analyze the abnormity of the multiplex telemetering signal,and utilize the traditional method of principal component analysis(PCA)to monitor the abnormity in an integrated way.In order to monitor the suspected abnormity,the monitoring should be further intensified to avoid omitting monitoring.And the wavelet analysis on the preprocessing of denoising on the data should meet the requirements of the spacecrafts to operate under complex environment and confirmatory experiment is being conducted aiming at related subsystem of the satellite.(2)The application of faults detection theory is of the key parts of the diagnosis system.Against the actual receipt signal,the top priority is to match and settle the various communication protocols of signals.Those signal data which are meeting the requirements of faults detection theory are the precondition of the faults detection of the system.The processing of signal should take speed and other factors into consideration and select the appropriate systematic solutions and design modes.(3)In terms of the establishment of the health management system in the spacecraft,the measures to detect faults and processing scheme are improved gradually.In the autonomous system management of space,the practical management strategy is being put forward with the combination of the present experience and other techniques like model analysis,thus establishing one intellectual treasure in terms of faults diction and health management of the satellite.