【24h】

Integrating KPCA with an Improved Evolutionary Algorithm for Knowledge Discovery in Fault Diagnosis

机译:将KPCA与改进的进化算法相集成,以进行故障诊断中的知识发现

获取原文
获取原文并翻译 | 示例

摘要

In this paper, a novel approach to knowledge discovery is proposed based on the integration of kernel principal component analysis (KPCA) with an improved evolutionary algorithm. KPCA is utilized to first transform the original sample space to a nonlinear feature space via the appropriate kernel function, and then perform principal component analysis (PCA). However, it remains an untouched problem to select the optimal kernel function. This paper addresses it by an improved evolutionary algorithm incorporated with Gauss mutation. The application in fault diagnosis shows that the integration of KPCA with evolutionary computation is effective and efficient to discover the optimal nonlinear feature transformation corresponding to the real-world operational data.
机译:在本文中,提出了一种新的知识发现方法,该方法基于内核主成分分析(KPCA)与改进的进化算法的集成。利用KPCA首先通过适当的核函数将原始样本空间转换为非线性特征空间,然后执行主成分分析(PCA)。但是,选择最佳内核功能仍然是一个未解决的问题。本文通过结合高斯变异的改进进化算法解决了这一问题。在故障诊断中的应用表明,将KPCA与进化计算相集成可以有效地发现与实际操作数据相对应的最佳非线性特征变换。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号