首页|Development of accelerated reliability test cycle for electric drive system based on vehicle operating data
Development of accelerated reliability test cycle for electric drive system based on vehicle operating data
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NSTL
Elsevier
For the current traditional vehicle, reliability test specification is difficult to effectively cover the actual service load intensity of the electric drive system (EDS), the existing specification has the problems of insufficient evaluation of extreme operating conditions, the inaccurate definition of the whole life cycle damage target, and weak user relevance. Therefore, this paper proposes a method to construct an accelerated reliability test cycle for EDS based on 300 users' one-year operation data. By extracting the running fragments and constructing the characteristic parameters associated with the dominant failure load of the EDS. Principal component analysis and clustering algorithms identify the running fragments and obtain five typical operating conditions. Based on the unit damage intensity distribution model, the accelerated testing condition for simultaneous assessment of multiple components is selected, and each component's whole life cycle damage target is determined. To further consider the loading sequence of the fragments, a state transfer probability matrix for each condition is established based on the Markov chain model, which implements the condition combination. Using the multi-objective optimization algorithm determines the number of cycles for each condition and eventually compiles the accelerated reliability test cycle of the EDS. In addition, by conducting rig tests to verify the validity of the accelerated test load spectrum and determine the accelerated factor by the testing data. This study provides specific references and technical support for developing and validating the EDS.
Electric drive system (EDS)Accelerated reliability testOperating data analysisCondition recognitionLoad spectrum compilationFATIGUE DAMAGELOADALGORITHMDESIGNMOTOR
NSTL
Elsevier
