机械结构的多目标模糊优化方法

Multiobjective fuzzy optimization of mechanical structures

结构设计的初始阶段含有大量的模糊信息，设计概念的产生、设计参数的取值、设计目标的评价、约束条件的允许范围以及工况等实际上都含有不同程度的模糊性提出了处理多目标模糊优化中目标与约束权重的向量水平截集法和权重最大最小法，拓展了向量水平截集法与最大最小法的应用范围研究了目标函数与约束函数的隶属函数的选取对优化结果的影响规律，在权重最大最小法中使用非线性隶属函数，更能有效地处理多目标优化中分目标的重要程度，以三杆辆架的最轻重量和最小变形的结构模糊优化问题为例证明了该方法的有效性。

There is a vast amount of fuzzy information in the primary stage of structural design, different degree of fuzziness included in generation of design concepts, range of the design variables, evaluation of the design objects ,allowable range of the constraints and load in practice. Vector tevel Cut Method and Weight Max - Min Method are proposed for weight of the objective functions and constraint functions easily, the ordinary Level Cut Method and Max - Min Method for symmetrical fuzzy optimization are extended. The effect of the membership functions of the objectives and constraints on the optimal result are investigated, importance of individual object is treated efficiently by nonlinear membership functions in Weight Max - Min Method. The effectiveness of these methods is illustrated with the fuzzy optimal design of a three - bar truss for minimum weight and hanimum deflection.