船舶柴油机机械磨损故障诊断的模式识别
|
摘要
船舶柴油机在工作过程中,经常会发生机械磨损故障,给船舶柴油机的工作稳定性带来困扰,针对当前船舶柴油机机械磨损故障存在的诊断准确率低、机械磨损故障诊断时间复杂度高等缺陷,设计了一种船舶柴油机机械磨损故障诊断的模式识别方法。首先分析当前船舶柴油机机械磨损故障的原理,并提取船舶柴油机机械磨损故障诊断特征,然后采用层次分析法分析确定每一个船舶柴油机机械磨损故障特征的权值,并根据RBF神经网络确定船舶柴油机机械磨损故障诊断的模式识别模型,最后进行船舶柴油机机械磨损故障诊断的验证性测试,分析本文方法的船舶柴油机机械磨损故障效果。本文方法的船舶柴油机机械磨损故障诊断率超过了90%,不仅远远高于对比方法的船舶柴油机机械磨损故障诊断率,而且船舶柴油机机械磨损故障效率得到有效的改善,具有很好的推广前景。
Marine diesel engine often suffers from mechanical wear faults during its working process, which brings troubles to the working stability of marine diesel engine. Aiming at the defects of low diagnostic accuracy and high time complexity of mechanical wear fault diagnosis of marine diesel engine, a pattern recognition method for mechanical wear fault diagnosis of marine diesel engine is designed. Firstly, the principle of the current marine diesel engine mechanical wear fault is analyzed, and the fault diagnosis features of marine diesel engine mechanical wear are extracted. Then, the weights of each marine diesel engine mechanical wear fault feature are determined by analytic hierarchy process(AHP). Based on RBF neural network, the pattern recognition model of marine diesel engine mechanical wear fault is determined. Finally, the marine diesel engine mechanical wear fault is analyzed. Finally, the validation test of the mechanical wear fault of marine diesel engine is carried out, and the effect of this method on the mechanical wear fault of marine diesel engine is analyzed.
Marine diesel engine often suffers from mechanical wear faults during its working process, which brings troubles to the working stability of marine diesel engine. Aiming at the defects of low diagnostic accuracy and high time complexity of mechanical wear fault diagnosis of marine diesel engine, a pattern recognition method for mechanical wear fault diagnosis of marine diesel engine is designed. Firstly, the principle of the current marine diesel engine mechanical wear fault is analyzed, and the fault diagnosis features of marine diesel engine mechanical wear are extracted. Then, the weights of each marine diesel engine mechanical wear fault feature are determined by analytic hierarchy process(AHP). Based on RBF neural network, the pattern recognition model of marine diesel engine mechanical wear fault is determined. Finally, the marine diesel engine mechanical wear fault is analyzed. Finally, the validation test of the mechanical wear fault of marine diesel engine is carried out, and the effect of this method on the mechanical wear fault of marine diesel engine is analyzed.
引文
[1]WIDODO A, YANG B S. Application of nonlinear feature extraction and support vector machines for fault diagnosis of induction motors[J]. Expert Systems with Applications, 2007,33(1):241–250.
[2]赵海洋,徐敏强,王金东.改进二叉树支持向量机及其故障诊断方法研[J].振动工程学报, 2013, 26(5):764–770.
[3]胥永刚,孟志鹏,陆明.基于双树复小波包变换和SVM的滚动轴承故障诊断方法[J].航空动力学报, 2014, 29(1):67–73.
[4]张建华,侯国莲,孙晓刚,等.采用概率神经网络的汽轮机故障诊断方法[J].动力工程, 2005, 25(5):698–701.
[2]赵海洋,徐敏强,王金东.改进二叉树支持向量机及其故障诊断方法研[J].振动工程学报, 2013, 26(5):764–770.
[3]胥永刚,孟志鹏,陆明.基于双树复小波包变换和SVM的滚动轴承故障诊断方法[J].航空动力学报, 2014, 29(1):67–73.
[4]张建华,侯国莲,孙晓刚,等.采用概率神经网络的汽轮机故障诊断方法[J].动力工程, 2005, 25(5):698–701.