基于改进鲸鱼算法寻优SVM的船用柴油机燃油系统故障诊断方法研究

黄烨鑫; 万振刚; 程  琛

基于改进鲸鱼算法寻优SVM的船用柴油机燃油系统故障诊断方法研究

引用本文：黄烨鑫，万振刚，程琛.基于改进鲸鱼算法寻优SVM的船用柴油机燃油系统故障诊断方法研究[J].计算技术与自动化,2021,(2):53-56

摘要点击次数: 474

全文下载次数: 0

作者	单位
黄烨鑫，万振刚，程琛	（江苏科技大学，江苏镇江 212003）

中文摘要:鉴于船用柴油机的复杂性，难以及时有效地进行维护保养决策，故此提出一种拉普拉斯分值和改进鲸鱼算法的支持向量机相结合的故障诊断方法。首先利用LS对征兆样本集进行降维处理，然后通过IWOA来优化SVM的惩罚因子和核参数，构造成分类器模型来进行故障诊断。将改进的算法与传统的算法进行比较，验证了改进鲸鱼算法寻优SVM在故障诊断方面的有效性。

中文关键词:船用柴油机拉普拉斯分值(LS) 改进鲸鱼算法(IWOA) 支持向量机(SVM) 故障诊断

Research on Fault Diagnosis Method for Marine Diesel Engine Fuel System Based on Improved Whale Algorithm Optimized SVM

Abstract:In view of the complexity of marine diesel engines, it is difficult to make timely and efficient maintenance decisions. Therefore, a Laplacian Score (LS) and an Improved Whale Optimization Algorithm (IWOA) Support Vector Machine (SVM). First, the LS is used to reduce the dimensionality of the symptom sample set, and then the penalty factor and kernel parameters of the SVM are optimized through IWOA, and a classifier model is constructed for fault diagnosis. The improved algorithm is compared with the traditional algorithm, and the effectiveness of the improved whale algorithm to optimize SVM in fault diagnosis is verified.

keywords:marine diesel engine Laplacian Score(LS) Improved Whale Optimization Algorithm(IWOA) Support Vector Machine(SVM) fault diagnosis

查看全文 查看/发表评论 下载pdf阅读器