Small Sample Reliability Synthesis Evaluation For The Engine/ISG Hybrid Power System Based On Bayes Theory

The characteristic of Small-sample system has little sample and little information. Based on the Probability Theory, Symbolic Stat., Random Process, the conventional reliability deducing method depended on large-sample data is impossible to apply and the reliability evaluation cannot be gone on wheel. The reliability evaluation of the Small-sample system has been one of the technique problems in the engineering practice. Based on an engine/ISG hybrid power system in Small-sample case system, the Bayes theory and application research of reliability evaluation method, which is applicable to Small-sample production system, is developed. After the engineering appliance analyses of engine/ISG hybrid power system, engineering practicability and validity of the theory method was validated. The reliability evaluation of the Small-sample system developed in the paper is applicable to engineering practice case. Then the ration and qualitative evidence are provided for the quality evaluation of the engine/ISG hybrid power system at the development stage. The main research content in this paper is an innovation of Bayes method and pyramid system reliability synthesis evaluation technique application to hybrid power system reliability, the research result is shown as follows:1) Based on the analysis of reliability information or failure mode effect and criticality, some important parameters or factors of system reliability are put forward. And reliability failure mode analysis is finished in combination with engine/ISG hybrid power system.2) System fuzzy reliability is established by the analysis of system failure state and fuzzy event probability, and then system reliability synthesis calculation formula is deduced in combination with system reliability factors. The formula establishes the theory basis for system reliability synthesis.3) Each variables value in system reliability synthesis formula above-mentioned is simulated randomly, and the simulation data is fitted with optimum distribution fitting formula and reliability prior probability density function of the hybrid system is determined.4) Based on the Bayes approximate evaluation method of system reliability, reliability point estimation and credible lower limit of the hybrid system are determined. And configuration reliability of engine/ISG hybrid system is simulated randomly and calculated to show the effect of system reliability information variables on system configuration reliability. Finally some parts failure of the hybrid system is analyzed.