CAJ | 학술논문

根据BP神经网络的非线性、强大的容错能力和具有泛化能力等特点，选择其作为故障诊断的工具。首先利用Multisim产生数据样本集，再根据电路结构确定神经网络结构，最后利用MATLAB完成神经网络的训练，计算出了隐层和输出层的权值和阈值。实验证明，BP神经网络对于调速控制器的故障诊断具有一定的实用价值。
근거BP신경망락적비선성、강대적용착능력화구유범화능력등특점，선택기작위고장진단적공구。수선이용Multisim산생수거양본집，재근거전로결구학정신경망락결구，최후이용MATLAB완성신경망락적훈련，계산출료은층화수출층적권치화역치。실험증명，BP신경망락대우조속공제기적고장진단구유일정적실용개치。
BP neural network based on the nonlinear, fault -tolerant capability and the characteristics of such a generalization, selecting it as a troubleshooting tool. Using sample set of data generated by Muhisim, according to the circuit structure of neural network structure, completion of the final use of MATLAB neural network training, to calculate the hidden layer and output layer weights and thresholds. Experiments show that the fault diagnosis for the speed controller has a certain practical value.