Diesel Multi-state System Reliability Analysis And Modeling

With the fast development of science and technology, the structrue of diesel engine system is more comlicated, and the working condition is severer. Therefore, the requirements of the product is more higher, the diesel should not only have good performance, but also a high level of reliability. Usually the research of diesel reliability, only two possible states were considered: completely reliable or complete failure, ignoring the influence of partial failure on the system performance, and the reliability analysis models are often quite different from the actual situation. The multi-state reliability analysis of diesel has great significance to improve the reliability level, availability, maintainability and security of diesel, prolong its life and make it more competitive. In this paper, finite element analysis theory, reliability theory and mathematical theory were used on the multi-state reliability analysis of the diesel engine. The main work as follows:(1)The crankshaft of diesel was analyzed by finite element analysis. In crankshaft working process, the high stress region mainly concentrates in the knuckles of the crank arm & the main journal, which was the area most easily broken. The finite element analysis method of Pro/E and ANSYS software was used to the stress analysis of crankshaft, then the stress value, displacement cloud and stress contour of crankshaft knuckle were obtained. And the improvement solutions were given.(2)In order to analysis the multi-state reliability of the diesel engine, the method of Markov model was proposed. The components and system in diesel engine were all analyzed, and the reliability function and instantaneous performance were obtained.(3)The universal generating function and the reliability block diagram were combined to analyze the more complicated reliability problem: multi-state system with uncovered failure in diesel engine. The reduced generating function and the integrated generating function in multi- state system were obtained, the reliability and performance indices were evaluated.(4)The universal generating function and Markov model were combined to do the reliabilily optimizaiton of series-parallel multi-state system under maintenance policies. The minimum cost configuration of the component was adopted and the system reliability demand was satisfied.