论文部分内容阅读
为有效提取往复泵工作时非平稳时变信号中的故障特征和将故障特征准确分类,提出以泵缸内的压力作为系统特征信号来提取故障特征向量的方法.将小波包分解的“频率-能量-故障识别”模式诊断方法引入泵阀工作状态监测中,通过改进的BP神经网络进行故障诊断.试验确定了网络的初始值,即选择学习率初始值为1.5、惯性因子为0.6、网络结构为3层的BP网络,其中隐含层的节点数为19个,即网络的结构是8-19-3.结果表明,该法降低了对原始信号处理的难度,且各阀箱内的压力之间无相互影响.该技术已应用于某船载系统的往复泵实时故障诊断中,实验验证了其有效性.
In order to extract the fault features of non-stationary time-varying signals effectively and classify the fault features accurately, a method of extracting the fault eigenvectors by using the pressure inside the pump cylinder as the system feature signal is proposed. - energy - fault identification "mode diagnosis method is introduced into the working condition monitoring of pump valve and the fault diagnosis is carried out by the improved BP neural network.The initial value of the network is determined by experiment, that is, the initial value of the selected learning rate is 1.5, the inertia factor is 0.6, The network structure is a 3-layer BP network, in which the number of hidden layer nodes is 19, that is, the network structure is 8-19-3. The results show that this method reduces the difficulty of processing the original signal, The pressure has no mutual influence.This technology has been applied to the real-time fault diagnosis of the reciprocating pump of a shipborne system, and its effectiveness has been verified experimentally.