Genetic Algorithms for Optimization of Complex Nonlinear System
暂无分享,去创建一个
Reasonable dimension reduction and effective optimization calculation are the basic ways to study the optimization of complex nonlinear system. A method based on the traditional decomposition technology is put forward to solve the optimization problem of complex nonlinear system, of which decision variable is decomposed to independence variable and coupling variable. On the basis of this method, genetic algorithms to solve the coupling variable and the traditional optimization technology to solve the independence variable are also established. This method has been used to a drainage optimal planning of a closed polder system. The result indicates that the method can obtain the optimum relation rapidly. It also can improve the calculation efficiency as well as avoid the difficultly to obtain the global optimal solution of the traditional optimization technology. This method can be applied to solve the similar optimization problem of other complex nonlinear systems.
[1] John R. Birge. A dantzig-wolfe decomposition variant equivalent to basis factorization , 1985 .
[2] Z. Michalewicz,et al. Genocop III: a co-evolutionary algorithm for numerical optimization problems with nonlinear constraints , 1995, Proceedings of 1995 IEEE International Conference on Evolutionary Computation.