A BP neural network based prediction method for electromagnetic compatibility

Abstract

In this paper, an artificial neural network based method is proposed in order to better evaluate and predict the electromagnetic compatibility. There are many inherent shortcomings in traditional BP neural networks. Thereafter, two improved BP neural network based method are introduced: BP neural networks with an additional momentum term and self-adaptive learning rate BP neural networks. The feasibility of applying these two methods to the evaluation and prediction of EMC. A simulation experiment is also given, together with the comparison of the two improved BP neural networks.