Reliability Robust Design Based On Uncertainty Analysis

With the development of electromechanical products in the direction of intelligence,large-scale and complex,the research on the uncertainties affecting the performance of electromechanical products has also entered a new stage.As an effective method of improving the quality of the product,robust design has been widely used in various fields.In order to adapt to the development trend of modern electromechanical products,based on the robust design theory,a variety of robust design optimization models under uncertain conditions are proposed for different engineering problems.The main contents are as follows:(1)After summarizing the research results of many scholars at home and abroad,through sorting out the development process of robust design theory,we have made clear the superiority of robust design in the problem of uncertainty optimization.The characteristics of robust design methods under different theoretical models and different engineering applications are analyzed.The advantages and disadvantages of each robust design model are summarized,and the future development direction is predicted and prospected.(2)In the current study of robust design,the indeterminate factors for simplifying the customary assumptions of the model are independent.However,in real engineering systems,the correlation between uncertain factors is objective and can not be ignored for the final optimization results.Therefore,it is very necessary to accurately depict the correlation between uncertain factors.In this paper,the Copula function is used to analyze the related uncertainties affecting the performance function.Secondly,a new method is proposed to solve the covariance between variables based on the Copula function.Finally,a robust design tolerance based on the Copula function to describe the correlation of uncertainties is established.Finally,the effectiveness of this method is proved by the design of the rack and pinion steering mechanism of the automobile.(3)Considering the characteristics of less information and more factors involved in the optimization design of complex electromechanical products,a robust design method based on sensitivity analysis for interval uncertainty is proposed.This method will focus on the selection of factors that have a greater impact on product performance,and appropriately ignore the uncertainty of the smaller factors while achieving robust results.Firstly,the engineering applicability of the global sensitivity analysis method Sobol' method and the interval uncertainty theory are discussed.Then,a robust design framework with interval uncertainty based on sensitivity analysis is constructed.Finally,an engineering example is used to verify and relevant research conclusions are given.(4)In order to focus on the future and effectively improve product design level,the treatment of time-varying and uncertain factors has become a key problem to be solved urgently.Taking into account the time-varying characteristics of the influence of various parts of the mechanical system on the overall performance,based on the existing sensitivity analysis of the interval,a time-varying interval sensitivity analysis method was proposed.Then,using the idea of reliability and robust design,a reliability robust design model based on time-varying interval sensitivity analysis was constructed.Firstly,based on the existing method of interval sensitivity analysis,the time variable is added and the method of interval sensitivity quantification under time-varying conditions is proposed.Secondly,by analyzing the time-varying interval sensitivity analysis method proposed in this paper,the three properties of this method are obtained and proved.Finally,a deduction is given,and a mathematical example is used to verify the properties and inferences.Subsequently,basing on the time-variable sensitivity method proposed in this chapter,and incorporating the idea of reliable and robust design,a robust robust design model based on time-varying interval sensitivity analysis was established.Finally,the relevant conclusions were given.