【摘要】 把实验设计中的均匀设计思想引入优化设计,提出了一种将确定性和随机性相结合的优化方法。该方法根据均匀设计原理在优化模型的设计变量空间内均匀分布一系列点,然后,将可行域内的上述系列布点作为优化计算的系列初始点,并选用常规模拟优化算法,分别开始进行优化计算,得到优化模型的一系列局部最优点。如果必要的话再在局部最优区加密网格,然后模拟额外的数据,那么能得到这些数据的改进最优解。并且重复插值过程该算法可求取非线性多峰函数的全局最优解。编制了计算程序,给出了计算实例,计算结果表明该设计方法是可行的。更多还原

【Abstract】 In this paper, some ideas of uniform design in the test design are introduced into optimization design and a new method of design combining deterministic method with random method is produced. A series of uniformly distributed points are generated by the principle of uniform design in variable design space. These points are regarded as series start points of the optimization model, and the conventional simulation optimize algorithms are selected to compute and optimize and a series of local optimal points are gained. If necessary, refined grids in local optimal space, then simulating extra datum, then the improved value of the optimal solve can be gained, and interpolating again and again, the global optimal solutions of nonlinear multi-peak function by this method. According to the method, a program is compiled and an example of design was implemented. The design results of the example testifies that the method is feasible.更多还原