BP Neural Network Trained by Strong Tracking Filter and It''s Application Investigation on System Identification
基于强跟踪滤波器训练的前向神经网络及其在系统辨识中的应用研究*

A new kind of training method of BP neural network is given,which is based on strong tracking filter.For the reason of forcing the remains orthogonality, it converges rapidly and the precision is higher compared with the Extended Kalman Filter(EKF). At last, an identification model example is given to show the effectiveness of the method.