关键词: 多目标优化, 灰色系统理论, 灵敏度设计, 可靠性稳健设计

Abstract:

Additional items of sensitivity of objective and constraint functions are generated through analyzing the corresponding functions objective functions and constraints. Two models of optimization design for reliability are proposed based on the additional items of sensitivity. According to the limitations of information for objective functions obtained from different evaluators, the objective functions and the evaluators' assessment based on gray system theory was allocated in a certain range. The expert assessment method is presented to calculate the weight coefficients of the objective functions. Reliability-based robust optimization design of cylindrical helix spring is achieved by using the optimization toolbox and symbolic toolbox of Matlab. Numerical examples prove that the proposed approach which provided the theoretic basis for the reliability design of the mechanical elements has a higher precision and reliability.

Key words: multi-objective optimization, grey system theory, sensitivity design, reliability-based robust design