论文部分内容阅读
本文根据响应面法的基本思想设计试验方案,根据动态系统参数识别与优化的要求,提出适用于动态系统随机仿真优化数值计算的算法RSM,研制了一个完整的自动运用响应面法进行随机仿真优化的数值计算方案,并用C++语言研制响应面法优化程序,以车辆悬挂参数的优化为例进行了数值分析.与穷举法和Matlab的非线性最优化方法的计算结果比较,响应面法在相同计算精度的情况下计算量明显减少,还提供响应与因素之间关系的重要信息,有利于数值试验和理论研究的深入进行.
According to the requirement of dynamic system parameter identification and optimization, this paper presents an RSM algorithm which is suitable for numerical simulation of stochastic simulation of dynamic system. A complete automatic response surface method is developed for stochastic simulation optimization , And the optimization program of response surface method was developed by using C ++ language.The numerical analysis was carried out with the optimization of vehicle suspension parameters as an example.Compared with the results of exhaustive method and Matlab nonlinear optimization method, Computational accuracy is significantly reduced in the case of calculation, but also provide important information on the relationship between the response and factors, is conducive to numerical experiments and theoretical research in-depth.