论文部分内容阅读
往复泵的正常工作是确保煤矿生产顺利进行的关键,由于工作环境恶劣,往复泵的故障诊断是非常重要的,因此,深入地研究了小波包分析和概率神经网络在往复泵故障诊断中的应用。分析了基于小波包的往复泵故障提取机理,设计了基于小波包分析的往复泵故障特征提取流程;构建了基于概率神经网络的往复泵的故障诊断模型,设计了概率神经网络的基本结构。对往复泵进行了故障诊断分析,仿真结果表明小波包和概率神经网络能够准确地获得故障诊断的类型。
The normal operation of the reciprocating pump is the key to ensure the smooth progress of the coal mine production. Due to the poor working environment, the reciprocating pump fault diagnosis is very important. Therefore, the application of the wavelet packet analysis and probabilistic neural network in the fault diagnosis of the reciprocating pump . Based on wavelet package analysis, the fault extraction mechanism of reciprocating pump is analyzed, and the fault extraction process of reciprocating pump based on wavelet packet analysis is designed. The fault diagnosis model of reciprocating pump based on probabilistic neural network is constructed and the basic structure of probabilistic neural network is designed. The reciprocating pump is diagnosed by fault diagnosis. The simulation results show that the wavelet packet and the probabilistic neural network can accurately obtain the type of fault diagnosis.