A Method for Analog Circuits Fault Diagnosis by Neural Network and Virtual Instruments

The values of analog circuits´ input and output signals and the component parameters are continuous, and meanwhile there are inevitable tolerance and non-linear components in analog circuits, therefore the presence of these factors increases complexity of the analog circuits fault diagnosis. RBF and BP neural network are two widely used feedforward neural networks, LabVIEW is a graphical programming language, which can provide users with a visual and convenient design environment. On the basis of RBF and BP neural network, the theory and method of analog circuits fault diagnosis based on ANN (Artificial Neural Network) are described. By the way of hybrid programming with LabVIEW and Matlab, the API for analog circuits fault diagnosis is established. Experimental simulation study is carried out, respectively, using RBF and BP neural network. From the fault diagnosis results, it can be seen that the method of combining ANN with LabVIEW can not only show the fault diagnosis results visually and directly, but also ensure a considerable diagnosis accuracy.