摘要
针对采煤机滚动轴承故障特征向量提取较困难、多分类效果不理想等问题,提出了基于HGWOMSVM的采煤机轴承故障诊断方法。对轴承故障信号进行小波降噪处理,利用经验模态分解算法对降噪后信号进行分解,并提取能量特征值,作为MSVM的训练集和测试集。采用MSVM进行故障状态识别,并用HGWO算法对MSVM的参数进行优化。试验结果表明,相比于GWO、GA和PSO优化MSVM模型,基于HGWO-MSVM的采煤机轴承故障诊断模型可明显提高故障识别精度和效率。
In view of problems of difficult extracting of fault feature vector and unsatisfactory multi-classification effect of shearer rolling bearing,a fault diagnosis method for rolling bearing of shearer based on HGWO-MSVM was proposed.The bearing fault signal is denoised by wavelet and decomposed by empirical mode decomposition algorithm,then energy characteristic value is extracted and used as training set and test set of MSVM.The MSVM is used to identify fault status and parameters of MSVM are optimized by HGWO algorithm.The experimental results show that the fault diagnosis model of shearer bearing based on HGWO-MSVM can obviously improve accuracy and efficiency of fault identification compared with GWO,GA and PSO optimization MSVM model.
作者
孙明波
马秋丽
张炎亮
雷俊辉
SUN Mingbo;MA Qiuli;ZHANG Yanliang;LEI Junhui(School of Management Engineering,Zhengzhou University,Zhengzhou 450001,China)
出处
《工矿自动化》
北大核心
2018年第3期81-86,共6页
Journal Of Mine Automation
基金
国家自然科学基金资助项目(71271194)
河南省基础与前沿技术研究项目(162300410073)
关键词
煤炭开采
采煤机
滚动轴承
故障诊断
经验模态分解
混合灰狼优化算法
多分类支持向量机
coal mining
shearer
rolling bearing
fault diagnosis
empirical mode decomposition
hybrid grey wolf optimization algorithm
multi-class support vector machine
