Fmeca For The Diesel Engine Reliability

Diesel engine,as the motive power for vehicles,engineering machinery,ships and power station units,has been playing an important role in the national economy.The reliability technology as an important security technology is of great significance to ensure the normal use of diesel engines.As far as the design of diesel engine is concerned,the lack of systematic reliability as a guide results in the relatively scattered and independent reliability work.Thus it is very difficult to truly improve the reliability level of the new diesel engine.Meanwhile, due to the lack of coordination in the scattered reliability work,a lot of unnecessary repeated work has been done,increasing the designers' amount of work and reducing the efficiency of reliability work.More importantly,due to the lack of the unified and effective management,the results worked out by the reliability work are difficult to gain the unanimous recognition of the staff from various departments,and thus the credibility is not high.In response to the status quo,this thesis,through the analysis and study of the crucial reliability work in the diesel engine's design stage,puts forward the FMECA-centered reliability framework,organically combining such reliability work as DFT,FMECA,reliability prediction and reliability allocation,etc.FMECA is the foundation and center of diesel engine's reliability analysis.In order to enhance the objective credibility of analyzing FMECA,this thesis introduces the expert weighting factor which will be confirmed by the means of fuzzy attribute hierarchical mode,targeting at experts' difference in experience and knowledge in the expert system.And aiming at FMECA scores listed in details which are hard to be expressed with precision,this thesis will adopt the fuzzy set to replace the specific scores.Meanwhile,it is only a single failure mode that is analyzed according to the traditional FMECA method.Therefore,this thesis employs the fuzzy inference method, considering the various failure modes of a single part and the influence of the combination of various failure modes,thus making a more comprehensive analysis of FMECA and drawing the results closer to the actual fact. As an important supplementary means for FMECA technology,the Fault Tree techniques play a tremendous role in enhancing the quality of analyzing FMECA and simplifying the workload of FMECA.In this thesis,aiming at the randomness and fuzziness of the diesel engine's failure data,the concept of fuzzy numbers will be introduced to analyze the dynamic failure number,using the fuzzy numbers to replace the failure rate which originally need to be determined and establishing the fuzzy Markov model,thus drawing a more objective and credible result.At the same time,for diesel engines are mechanical products whose components' failure distribution are Weibull distribution,this thesis will adopt the failure rate of Weibull distribution instead of the failure rate in the original Markov chain complying with the exponential distribution,and employ the Markov model based on the Weibull distribution,so that the results of the analysis are more real-time and more in line with the actual projects.Focusing on FMECA,this thesis proposes the methods for reliability prediction and reliability allocation based on FMECA,and links with the results of FMECA work as the basis for prediction and allocation,thus organically combining FMECA, reliability prediction and reliability allocation in the whole design phase of diesel engines.