基于响应面法的超高压大口径闸阀阀体多目标优化

Multi-objective Optimization of Ultrahigh Pressure and Large Diameter Gate Valve Body Based on Response Surface Method

以某超高压大口径核级闸阀为研究对象,对其全开状态下的受力进行有限元分析,应力云图显示最大等效应力大于材料许用应力。且因其质量过大,装运困难,故对闸阀阀体进行多目标优化。选择5个输入变量和3个目标函数,对筛选出的敏感度较高的设计变量与其对应的目标函数构建响应面代理模型,耦合MOGA遗传算法得到优化结果。结果显示:最大等效应力降低了29.3%,阀体质量下降了5.5%,最大变形量减小了0.00052 mm。

Taking an ultrahigh-pressure large-diameter gate valve of nuclear grade as the object of research,the finite element analysis of its force under the fully-open state was implemented.The stress cloud diagram shows that,the maximum equivalent stress is greater than the allowable stress of the material.The large mass makes it difficult to ship and multi-objective optimization of its body becomes necessary.Having five input variables and three objective functions selected to construct response surface surrogate model for the selected design variables with higher sensitivity and their corresponding objective functions,and through coupling the MOGA genetic algorithm,the optimization results were obtained.The results show that,the maximum equivalent stress can be reduced by 29.3%,the body weight reduced by 5.5%and the maximum deformation reduced by 0.00052 mm.