Correlation Analysis And Research On The Telemetry Data Of Synchronous Satellites

With the vigorous development of China's aerospace industry, the number of satellites in-orbit increases year by year. Thus, safe and stable operation of the satellite, timely detection and treatment of satellite failures, have become the focus of management of the satellites in-orbit. During the process of satellites on-orbit and in-orbit monitoring, large amounts of information will be produced, including information of fault, status mutation information, monitoring and computing results, record of the satellite's working status, daily management information, and environmental parameters of the operational space. These telemetry data is more realistic than overground simulation information and testing data, directly reflecting the state of the satellite payloads and satellite operation. Thus, these data is the main contents of satellite monitoring and state analysis.It is rich resources and should be fully analyzed and effectively used.The long-term monitoring information indicates that, there is a certain correlation between the satellite status and operating information. The abnormal operation of a satellite is often accompanied by the mutations or slow change of the satellite status. The result of correlation analysis between the satellite staus and operational information can be used to analyze the performance of the satellite in-orbit, to timely detect the abnormal performance, as well as to exclude the faults. These jobs have done provide a guarantee for the organization of satellite management activities and completing the given tasks. Meanwhile, the analysis results have significant meaning to the development of the satellites'potentials and the forwarding satellite design.Furtherly understanding the basic idea and mining process of ID3 algorithm, the advantages and disadvantages of the algorithm are found combined with practical analysis process of telemetry data. Optimization improvements are made for its shortcomings in practical application. For the disadvantage of the algorithms tends to choose the property which has more value as a splitting attribute, the optimization algorithm selects splitting attribute on the basis of information gain ratio; in addition, for the continuous data, it is difficult for the algorithm to analyze the continuous data. In order to overcome this shortcoming, the continuous data is divided into several sections in the optimization algorithm. Through the comparative analysis and evaluation of the mining result, we find that the classification accuracy of the improved algorithm is relatively higher, suiting the needs of satellite engineering better. Based on the improved ID3 algorithm, the correlation analysis system of the telemetry data is designed, achieving the mining analysis of the solar panels' telemetry data.In this paper, the improved ID3 algorithm is used to analyze the correlation of the telemetry data of solar panels.The key influencing factors and the correlation during in factors of the solar panels are intuitivly displayed by constructing decision dree.The anlysis results also provide the decision basis for failure analysis of the solar panels.The targeted analysis of failures can reduce the blindness of analysis. We can improve the working efficiency. As a result, it is significant for analyzing the performance and status of solar panels during the satellite in-orbit operation process.