最小偏差法在机械多目标优化设计中的应用  

Application of Minimum-Deviation Method to Mechanical Multi-objective Optimal Design

作　　者：魏锋涛[1] 宋俐[1] 李言[1] 

机构地区：[1]西安理工大学机械与精密仪器工程学院,陕西西安710048

出　　处：《工程图学学报》2011年第3期100-104,共5页Journal of Engineering Graphics

基　　金：陕西省重点学科建设专项资金资助项目(102-00X903);西安理工大学科学研究基金资助项目(102-210913)

摘　　要：提出了处理多目标问题的最小偏差法,并将其应用于机械多目标优化设计。以流体动压滑动轴承多目标优化设计为例,建立了以摩擦系数最小、发热量最小和承载能力最大为目标函数的多目标优化设计数学模型,根据最小偏差法原理构造统一目标函数,利用改进遗传算法对该问题进行了优化设计。算例整个求解过程和结果表明,该方法既可以避免人为因素的影响,又能够获得比常规设计更佳的设计参数,同时也验证了所提方法对于解决机械多目标优化设计问题的有效性和可行性。The minimum-deviation method is introduced into the mechanical multi-objective optimal design.Taking a multi-objective optimal design of hydrodynamic sliding bearing as the example and taking the minimization of friction coefficient and heat productivity,and the maximization of load capability as objectives,the minimum-deviation method is adopted to establish the multi-objective optimal mathematical model.The improved genetic algorithm is utilized to solve this multi-objective optimal problem.The solution-seeking process and its results show that the minimum-deviation method can obtain better results than conventional methods in dealing with multi-objective optimal problem and the validity and feasibility of the method is verified.

关 键 词：最小偏差法 机械多目标 优化设计 改进遗传算法 

分 类 号：TH132.44[机械工程—机械制造及自动化]