一种基于改进遗传规划的泛函网络结构逼近模型

摘要

The concept of generalized function is put forward based on the structure characteristics of functional network and the global search ability of genetic programming, the generalized function is studied through the improved genetic pro gramming encoding,the fitness function is designed using least-square method,and functional network structure model of the optimal approximation is gotten.Finally,four numerical simulation examples indicate that this method is effective and feasible, and has strong generalization.%基于泛函网络的结构特点和遗传规划的全局搜索能力,提出了广义基函数概念,通过改进遗传规划的编码方式对广义基函数进行学习,用最小二乘法设计适应度函数,从而确定泛函网络的最佳逼近结构模型.最后,4个数值仿真实例表明,该方法是有效可行的,具有较强的泛化特性.