Research on the Fault Diagnosis of Analog Circuits Based on the Data Fusion with Neural Network
Yigang He · Journal of Hunan University · 2005
According to the principle of fault detection and diagnosis, the technology of data fusion with neural network was used to deal with a lot of data obtained from fault detection and diagnosis. A new method based on Dempster-Shafer theory of evidence to solve fault detection and diagnosis was proposed. These procedures could be implemented in a multi-layer neural network with specific architecture consisting of one input layer, two hidden layers and one output layer. The weight vector, the receptive field and the class membership of each prototype were determined by minimizing the mean squared differences between the classifier outputs and the target values. The DS-based neural network actually modified Radial Basis Function Network (RBF). Experiments with simulated and real data demonstrated the excellent performance of the proposed fault diagnosis approach.