摘要
本文介绍了一种遗传算法(GA)优化自适应神经模糊推理系统(ANFIS)的方法,并采用基于GA优化ANFIS方法,拟合非线性多峰函数,同时分析了这种方法的拟合能力和预测能力。实验结果表明,加入GA优化后的ANFIS具有更加优秀的拟合能力和预测能力,更适合于用来建立复杂参数间的非线性映射关系。
This article introduces a method that Adaptive Neural-based Fuzzy Inference System (ANFIS) optimized by Genetic Algorithm (GA).Using the method fited more non-linear function. And based this,I analyses the method's ability of fitting and prediction . The results of simulation show the ANFIS Based on GA has a better performance than ANFIS on function fitting and prediction.ANFIS Based on GA has higher precision than ANFIS. It is more suitable for creating the non-linear mapping between complex interrelationship parameters.
出处
《微计算机信息》
2009年第3期211-212,75,共3页
Control & Automation
作者简介
丁浩(1982-),男(汉族),中国地质大学(武汉)计算机学院硕士研究生,主要研究方向:机器学习、数据挖掘。通讯地址:(430074湖北省武汉市中国地质大学2007级硕研11班)