实际工程优化问题通常涉及多个目标,这些目标往往不能显式表达从而加大了优化的难度。智能多目标优化方法可以在不考虑问题具体特征的前提下进行优化解的搜寻,但需要多次的目标值计算,限制了在求解耗时的问题中的应用。常见的基于代理模型的方法能较好地处理效率问题而成为研究热点,但面临着求解精度较低的问题。鉴于此,本文针对基于代理模 The actual project optimization problems usually involve multiple goals, these goals often can not be explicitly expressed thereby increasing the difficulty of optimization. The intelligent multi-objective optimization method can search for the optimal solution without considering the specific characteristics of the problem, but needs multiple target values ​​calculation, which limits the application in solving the time-consuming problem. Common methods based on agent model can deal with the efficiency problem well and become a research hotspot, but they face the problem of low precision. In view of this, this article is based on agent model