基于Elman神经网络的传感器故障诊断研究

针对传统的传感器故障诊断技术的不足,提出一种基于Elman神经网络的故障诊断方法,建立了Elman网络故障诊断模型,利用小波包分解方法获取用于训练神经网络的特征能量谱,对所建立的模型进行训练。为了检验模型的实际诊断能力,以某动力系统管路流量传感器的4种典型故障诊断为例进行仿真实验,并和标准BP神经网络的诊断结果进行对比。仿真结果表明:基于Elman神经网络的故障诊断速度更快、准确率更高、泛化能力更强,验证了所提出方法的实用性和有效性。

Study of sensor fault diagnosis method based on Elman neural network

In view of disadvantages of traditional sensor fault diagnosis techniques,a fault diagnosis method based on Elman neural network is proposed and a diagnosis model based on Elman neural network is established.The characteristic energy spectrum which is used in network training is obtained by wavelet packet decomposition method and it is used in the training of the established model.To test the practical diagnosis ability of the model,4 kinds of typical fault diagnosis of a sensor of power system piping flow are taken as examples to conduct a simulation test.The diagnosis result of the model based on Elman neural network is compared with the result of standard BP neural network.The simulation result shows that the method based on Elman neural network has a faster diagnosis speed,higher accuracy and stronger generalization ability.The method proposed in this paper is verified to be practical and effective.