The Information Fusion of Multi-sensor of Based on Federated Kalman Filter and Neural Networks

A new method of fault diagnosis is proposed. This method is called the complex fault diagnosis of based on federated kalman filter (FKF) and neural network (NN). It uses Kalman filter to estimate the measurable parameters´ variations of car engine multi-sensor, and processes fault signal by information reconstruction, and then trains and corrects noise error by neural networks. According to these, it can diagnoses automobile engine fault. The simulation results show that the method is feasible and effective.