Complex Product Reliability Measurement And Modeling Techniques

The satellite is one of the important spaceflight equipments. For a satellite, the reliability level is the most important index of the quality. To control the reliability of the satellite product, the paper aims to build a reasonable model to value the reliability of the satellite, and summarizes the reliability indexes of that. In the paper, it introduces many methods of modeling and valuing the satellite reliability, and gives the solving ways to describe the complex redundancy structures and Phased-Mission Systems (PMS) in the reliability analysis of the satellite.At first, the paper introduces the choice of reliability indexes. It researches the active standard of satellite indexes (GJB1909.4), and investigates the practice circumstance. Some primary indexes were summarized, and some problems in the implement were found out also. It brings the way to choose the proper indexes.Assuring the high mission reliability of the satellite, there are many complex redundancy structures. The paper sums up the representative structures, consisting of series-connection structure, parallel-connection structure, cold spare structure, hot spare structure, and ring spare structure, etc. For each structure, it researches the proper methods and approaches to modeling the reliability. The appropriate reliability models can analyze and value the system exactly, and find out the parts of weakness. It can supply the warrant to mend the design.Dynamic Fault Tree (DFT) method can describe the dynamic performance of satellite system. The processes of modeling are easy, and the structure of model is clear. So, DFT is used widely. The paper summarizes the modeling processes and disposing methods. In the traditional Dynamical Fault Tree analysis, it is necessary to modularize DFT tree firstly so as to obtain static subtrees and dynamic subtrees. Generally, Binary Decision Diagram (BDD) and Markov chains are utilized in the DFT to process static and dynamic subtrees, respectively. However, due to the possibility of state combinatorial explosion problem in Markov chain, it is difficult to analyze system with DFT in some cases. The paper investigates some other methods in DFT in order to solve this problem, consisting of trapezoid formula, Petri net, and bayesian network, etc. some example are analyzed with these new methods. The analysis results show that the proposed methods can overcome the state combinatorial explosion problem and guarantee high accuracy.At last, the paper researches the processing method of the PMS that the satellite is one of. The modeling of PMS is more complicated than that of single mission system. The phases of the satellite working have different states and request of reliability. At every phase, it must have individual reliability model to analyze. But the part state at the beginning of later phase is the same as that at the end of this phase. To solve this problem, the paper introduces a probability equivalent method. The proposed method can model and account PMS effectively.