摘要
为获得多峰值函数优化的多个极值,提高智能优化方法,获得多峰值函数优化解的精度,受生物免疫系统相关机理的启发,提出了基于工程混合免疫计算的多峰值函数优化方法,并给出了具体的求解算法.采用传统的蚁群优化方法以及本文方法对3个典型的多峰值复杂测试函数进行性能测试,对比每种方法的搜索代数、搜索到的峰值个数、最大适应度值以及平均适应度值.比较结果表明,本文方法具有增量学习能力且在分类准确率方面都高于传统的蚁群优化方法.
Inspired by the relevant mechanism of biological immune system,multi-modal function optimization based on a project hybrid immune algorithm was presented and a concrete algorithm was given to obtain multi-extreme points for multi-modal function optimization and to improve the accuracy of multi-modal function optimization given by intelligent optimization methods.To compare the solving performance of the classical ant colony optimization and the multi-modal function optimization based on a project hybrid immune algorithm,two methods were used to solve three complex testing functions.Their performances were compared in terms of searching times,the searched extreme number,the maximum fitness value and the average fitness value.Comparison results indicated that the given algorithm had incremental learning capacity and better classification accuracy than that of the classical ant colony optimization.
出处
《信阳师范学院学报（自然科学版）》
CAS
北大核心
2013年第1期140-142,150,共4页
Journal of Xinyang Normal University(Natural Science Edition)
基金
河南省自然科学基金项目(122300410310)
关键词
人工免疫
多峰值函数
优化
混合免疫算法
artificial immune
multi-modal function
optimization
hybrid immune algorithm
