Neural Network Integration Fusion Model and Application
Xiaodan Zhang, Feng Tian, Yuan Mu, Peigang Sun, Hai Yan Zhao · 2007
A new fusion model is proposed, which is the combination of BP neural networks and D-S evidence reasoning, to solve the problems of low precision rate in automotive engine fault diagnosis by traditional expert system. The method realizes feature level fusion of all subjective data and expert experiments on different parts of engine, and the predominance compensation of different models. In simulation experiment, this method proposed in this paper can improve diagnosis precision 5.0% more than expert system.