基于超限学习机的柱塞泵故障诊断
|
摘要
为了提高柱塞泵故障诊断的准确度,提出应用一种新的分类算法即超限学习机(ELM)对柱塞泵进行故障诊断与识别。首先,采集柱塞泵各工作状态下的信号,其次对采集到的信号进行预处理,选择出特征向量。最后,应用ELM与其它分类器分别对其进行诊断与识别。对比试验结果表明,新的方法故障诊断准确度高且诊断速度快。
In order to improve the accuracy of fault diagnosis of plunger pump, a new classification algorithm, the Extended Learning Machine(ELM), is proposed to diagnose and identify the fault of plunger pump. Firstly, the signals of the plunger pump in different working conditions are collected. Secondly, the collected signals are preprocessed and the eigenvectors are selected. Finally, ELM and other classifiers are used to diagnose and identify them respectively. The comparison test results show that the new method has high accuracy and fast diagnosis speed.
In order to improve the accuracy of fault diagnosis of plunger pump, a new classification algorithm, the Extended Learning Machine(ELM), is proposed to diagnose and identify the fault of plunger pump. Firstly, the signals of the plunger pump in different working conditions are collected. Secondly, the collected signals are preprocessed and the eigenvectors are selected. Finally, ELM and other classifiers are used to diagnose and identify them respectively. The comparison test results show that the new method has high accuracy and fast diagnosis speed.
引文
[1]刘思远,王闯,杨梦雪,等.斜盘泵滑靴副剧烈磨损过程的动态特性分析[J].液压与气动, 2016:1.
[2]吴胜强,姜万录.基于证据理论多源多特征融合的柱塞泵故障诊断方法[J].中国工程机械学报, 2011, 09(1):98-102..
[3] Huang G B, Zhu Q Y, Siew C K. Extreme learning machine:Theory and applications[J]. Neurocomputing, 2006, 70(1-3):489-501.
[4] Huang G B. An Insight into Extreme Learning Machines:Random Neurons, Random Features and Kernels[J]. Cognitive Computation, 2014, 6(3):376-390.
[5]吴胜强.核主元分析及证据理论的多域特征故障诊断新方法研究[D].燕山大学, 2011.
[2]吴胜强,姜万录.基于证据理论多源多特征融合的柱塞泵故障诊断方法[J].中国工程机械学报, 2011, 09(1):98-102..
[3] Huang G B, Zhu Q Y, Siew C K. Extreme learning machine:Theory and applications[J]. Neurocomputing, 2006, 70(1-3):489-501.
[4] Huang G B. An Insight into Extreme Learning Machines:Random Neurons, Random Features and Kernels[J]. Cognitive Computation, 2014, 6(3):376-390.
[5]吴胜强.核主元分析及证据理论的多域特征故障诊断新方法研究[D].燕山大学, 2011.