Fault diagnosis on analog circuits based on Integrated Learning Method

The Integrated Learning Method (ILM) uses multiple learners to solve the same problem, which can greatly improve the generalization ability of learning systems. To address the fault diagnosis on analog circuits, aiming at the shortcomings of diagnosis and model stability with single RBF neural network to diagnose faults of analog circuit system, the paper discussed method to improve model diagnosis accuracy with Bagging algorithm of ILM to integrated multiple neural networks. The experiment results show the adoption of this scheme can significantly improve the performance of neural network diagnostic model.