Reliability Evaluation Model Based On Multiple Parameters Degradation Correlation Failure For Accelerated Degradation Test

More and more products have a high reliability and long life because of the rapid development of technology,thus evaluation the reliability of long life products has become an important issue.However,in the relatively short term,long life products are almost impossible to fail,and even a "zero failure" phenomenon,which is difficult to assess the reliability of the product based on failure data.The Accelerated Degradation Test(ADT)is an effective way to evaluate the reliability of the product using the degraded data of the product,and to predict the reliability of the product.However,most of the researches on accelerated degradation tests are limited to single performance parameters.In practice,there are many failure mechanisms in the product,that is,there are multiple performance parameters degenerated at the same time,and the degradation data of multiple performance parameters exist correlation.Therefore,this paper considers the data of multivariate performance degradation and its correlation.Based on the Copula function and the statistical analysis evaluate the reliability of the product in the accelerated degradation test.Firstly,studied the research background and significance of the topic selection,the development process of the reliability test method,the research status of the accelerated life test and the accelerated degradation test at home and abroad,the research status of the statistical analysis method of the accelerated degradation test data.Secondly,the related theories of Copula function(such as the definition of Copula function,Sklar theorem,Copula function selection criteria and parameter estimation)are introduced.The theoretical basis of the reliability model of multivariate accelerated degradation test is established.Considering the product reliability evaluation problem of multiple performance parameters,the reliability model of the product degradation degree based on the degenerate quantity of the uniform distribution parameters of different distribution is established,and the reliability model of the multiple accelerated degradation test is established,including the reliability model of the multiple accelerated degradation test under the independent failure,the collection failure and the non-competition relation.Based on Copula function,the product reliability evaluation model based on the correlation test of multi-performance parameters is established.Based on the theory of mathematical statistics and Copula function,the reliability evaluation model of multi-critical performance parameter failure based on accelerated degradation test is established,including matching the trajectory of the degradation of the performance parameters;The deduction of the degenerate distribution of the performance parameters of the multiple accelerated degradation test;Infer the correlation structure between the degradation of multi-performance parameters;Infer the correlation between the degradation of multi-performance parameters.Considering the reliability evaluation of the product with multi-performance parameters under non-competition relation,the reliability evaluation models of multiple accelerated degradation tests under non-competitive relation is established.