Research On Composite Life Distribution Based On Chen's Distribution And Statistical Analysis

With the epoch development and the progress of science and technology,the functions of products have become plentiful,and people's demands for quality have become higher and higher.With the in-depth study of reliability analysis theory and the comprehensive development and application of the academic discipline,objects of study have gradually being expanded from cutting-edge industrial sectors such as aeronautics and nuclear power to motor and power systems,machinery and equipment,power,civil engineering,metallurgy,and chemical industries,this has an important impact on improving product quality and reliability.Failure data appearing in a wide range of practical processes in various fields presents skewed characteristics.Based on the study of such data,the academic community has emerged as the research direction of composite life distribution.Compared with the traditional foundation life distribution,the density function image features of the composite life distribution are more abundant,and the image characteristics of the failure efficiency function are more flexible and variable,so that the failure modes of different products can be better represented.In 2000,a two-parameter life distribution proposed by Chen Z,the failure rate function presents a bathtub or monotonous increasing shape feature,which can better adapt to the failure rate curve of actual life data.Based on this distribution and related research,this paper proposes two new composite lifetime distributions(PoissonChen distribution PC(?,?),Geometric-Chen distribution GC(?,?,p)),and the two new composite lifetime distributions are studied:(1)The properties and image characteristics of the density function and failure rate function of two new life distributions are studied,and the existences of two distribution origin moments are proved,and numerical values of the origin moments under different parameters are calculated numerically.(2)The maximum likelihood estimations of the parameters of the two new life spans under full sample data are studied,and Monte-Carlo simulation is performed to investigate the accuracy of the estimations;base on the Geometric-Chen distribution and the actual life data of the hydraulic pump,the maximum likelihood estimation of the parameters is obtained,the life data of the hydraulic pump obeys the GeometricChen distribution using the K-S fitting test.