文献类型：期刊文献

中文题名：基于免疫优势克隆文化算法的故障诊断方法及其在TE过程中的应用

英文题名：Fault Diagnosis Method Based on Immunodominance Clone Cultural Algorithm and its Application in TE Process

作者：王雪洁[1];黄海燕[1]

机构：[1]华东理工大学信息科学与工程学院,上海200237

年份：2015

卷号：42

期号：5

起止页码：255

中文期刊名：计算机科学

外文期刊名：Computer Science

收录：CSTPCD;;北大核心:【北大核心2014】；CSCD:【CSCD_E2015_2016】；

语种：中文

中文关键词：免疫优势克隆选择算法;文化算法;支持向量机;故障诊断

外文关键词：Immunodominance clone select algorithm, Cultural algorithm, SVM, Fault diagnosis

摘要：免疫优势克隆选择算法是一种新型的免疫算法,具有较强的局部和全局搜索能力。将其与文化算法结合,提出一种新型的免疫优势克隆文化算法,它可以更好地利用先验知识指导种群进化;并设计了新的动态接受函数来促进文化算法内部知识更新,提高算法的搜索能力。将该算法用于支持向量分类器的核参数优化中,构造性能良好的分类器,并将其用于Wine dataset的数据分类和化工TE过程的故障诊断中,实验结果表明,该算法能够准确地对SVM的核函数参数进行寻优,提高了故障诊断的准确性,具有应用推广价值。
Immunodominance clone select algorithm is a new kind of immunodominance algorithm with strong local and global search ability. Combining it with cultural algorithm, we put forward a new immunodominance clone cultural algorithm. It can make use of transcendental knowledge to guide the evolution of species. We designed a new dynamic acceptance function to promote knowledge updating in cultural algorithm and increase the search capability of the algorithm. Immunodominance clone cultural algorithm was used in nuclear parameter optimization of support vector machine （SVM） for constructing the good performance classifier. The results of Wine dataset classification and fault diagnosis of TE reflect that imrnunodominance clone cultural algorithm can search the nuclear parameter optimization of SVM accurately. It raises the accuracy of fault diagnosis and has better spread value.