Evaluation Method Of Engine Reliability Risk Based On Fuzzy Theory

With the fast development of science and technology, the structure of engine system is more complicated, the capacity parameter is constantly improved, and the working condition is severer, consequently the problems of operation security and reliability are increasingly prominent, hence the probabilistic risk assessment on engine's reliability become one of important means to improve its security and reliability as well as to control risk and reduce resource and expense of maintenance on fault. The main work of this paper is as follows:(1)The main bearing seat of engine is analyzed by finite element analysis. Based on the crack faults of main bearing seat and main bearing shell in engine, their models of finite element analysis are built by analyzing their multiple working conditions with the finite element analysis, thus the structure intensity and deformation value of each component can be obtained, finally the fault causes will be found and the corresponding solutions can be proposed.(2)The method of fuzzy state allocation on the fuel oil supply system of engine is proposed. Based on the fuzzy representation of each state about the fuel oil supply system of engine is given through fuzzy set theory, and the function eigenvalue as well as the feasible domain of fuzzy state about system is defined, then the feasible state space of system can be built. Once the state of each component is determined by function eigenvalue, the equivalence set can be established by low boundary point to describe the change of system, thus the method of establishing equivalence set is formed; finally the state of fuel oil supply system about engine can be assigned and forecasted by providing the general process and specific requirement of state allocation on multi-state system and establishing the optimization model of state allocation on system.(3)The probabilistic risk assessment method on engine's reliability is proposed by synthetically using fuzzy set theory, failure mode and effect analysis (FMEA) and fault tree analysis (FTA). The priority rank of failure modes is evaluated by fuzzy risk priority number (FRPN) in the fuzzy failure mode and effect analysis, where occurrence probability ranking(OPR), effect severity ranking(ESR) and detection difficulty ranking(DDR) are all fuzzy weighted geometric mean, and the FRPN can be computed using alpha-level set and optimization model. According to the risk order obtained from results above, a new method of fuzzy fault tree analysis is proposed to analyze these failure modes in sequence, thus the fuzzy probability of top event and the probability importance index of each basic event under diffirent alpha-level set can be obtained.