期刊文献+

BP神经网络在负载管理中心故障诊断中的应用

Application of BP Neural Network in Electrical Load Management Center Fault Diagnosis
下载PDF
导出
摘要 在分析电气负载管理中心故障特点的基础上,利用神经网络权值和阈值能够随实际的排故结果不断更新及正向推理速度较快的特性,提出了基于BP神经网络的负载管理中心故障诊断方案,并确立了故障诊断BP网络模型。借助于MAT-LAB的神经网络工具箱,采用两种改进的训练算法对网络进行训练,得到了用于诊断的BP神经网络模型,为检验该模型故障诊断的准确性,采用大量的数据样本进行了仿真。结果表明:基于神经网络的诊断方法故障识别率高、快速有效,具有良好的实用价值。 As the weight value and threshold value of neural networks could be updated with actual fault results and the forward inference speed is very quick, then on the basis of analysing the fault symptom of Electrical Load Management Center, a fault diagnosis scheme based on neural networks for ELMC is presented in this paper. By u- sing neural toolbox of MATLAB and adopting two kinds of improved neural network algorithms , fault diagnosis BP neural networks model was achieved. In order to testify the validity of the model, lots of simulations were carried out. Results of experiments show that this scheme can greatly improve the accuracy and speed of fault diagnosis. So this scheme has great application value.
作者 周素莹 林辉
出处 《计算机仿真》 CSCD 2005年第11期169-171,208,共4页 Computer Simulation
关键词 神经网络 负载管理中心 故障诊断 动量批梯度下降法 有弹回的算法 Neural network Electrical load management center(ELMC) Fault diagnosis Batch gradient descent with momentum algorithm Resilient back - propagation algorithm
  • 相关文献

参考文献6

二级参考文献8

共引文献57

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部