| BP神经网络在轴向柱塞泵故障诊断中的应用 |
| |
| 引用本文： | 高兴培.BP神经网络在轴向柱塞泵故障诊断中的应用[J].广东广播电视大学学报,2010,19(3):97-99,104. |
| |
| 作者姓名： | 高兴培 |
| |
| 作者单位： | 茂名广播电视大学,广东茂名,525000 |
| |
| 摘 要： | 针对轴向柱塞泵故障机理的复杂性和故障信息的不确定性，提出了基于人工神经网络的故障诊断方法。以一种典型设备的几种主要故障为例，设计了适合于故障诊断的BP神经网络模型，运用Levenberg-Marquardt优化算法进行网络训练，并针对网络训练中可能出现的过拟合、局部小、隐层节点数确定等问题制定了相应的网络优化策略，以保证训练后的网络具有较好的记忆和归纳能力，并用Vc＋＋6.0语言和SQL Server2000数据库开发了基于BP神经网络的轴向柱塞泵故障诊断系统，结果表明，该系统有良好的故障诊断精度和较强的泛化厶匕力。
|
| 关 键 词： | BP神经网络 Levenberg-Marquardt算法 轴向柱塞泵 故障诊断 |
Application of BP Neural Network in Axial Plunger Pump Fault Diagnosis |
| |
| GAO Xing-pei.Application of BP Neural Network in Axial Plunger Pump Fault Diagnosis[J].Journal of Guangdong Radio & Television University,2010,19(3):97-99,104. |
| |
| Authors: | GAO Xing-pei |
| |
| Institution: | GAO Xing-pei (Maoming Radio and TV University, Maoming, Guangdong, China, 525000) |
| |
| Abstract: | Considering complexity of mechanism and uncertainty of information in the axial plunger pump faults, a fault diagnosis method based on artificial neural network is discussed. According to a few fault of typical equipment, a fault diagnosis system model based on BP neural network is designed. Then Levenberg-Marquardt algorithm is used to train the neural network. In order to resolve some problem, such as overfitting, local minimum, the number of the hidden layer in the process of training, some practical strategies and methods are taken out to reduce the training time and to improve the ability of memory and epagoge. Finally a fault diagnosis system based on BP neural network is created with VC＋＋ 6.0 program language and SQL Server 2000 database. The experimental results show that this system has better precision and strong generalization ability. |
| |
| Keywords: | BP neural network Levenberg-Marquardt algorithm axial plunger pump fault diagnosis |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
