论文信息 - An Improved Neural Network Model and Its Applications

An Improved Neural Network Model and Its Applications

Elman Neural Network has been an efficient system identification tool in many areas. However, one of the problems often associated with this type of network is that the speed of learning is too slow. The HF Elman neural network is presented for the modeling of unknown delay and high-order nonlinear system. Then chaos searching is imported to train it, make BP algorithm can skip the local minimum and find the global minimum easily. The simulation result shows that the proposed method can speed up the original ENN algorithm and get good results for the prediction tasks.

Rongjun Li | Zhihang Tang

[1] Yanchun Liang,et al. IMPROVED ELMAN NETWORKS AND APPLICATIONS FOR CONTROLLING ULTRASONIC MOTORS , 2004, Appl. Artif. Intell..

[2] Stefan C. Kremer,et al. On the computational power of Elman-style recurrent networks , 1995, IEEE Trans. Neural Networks.

[3] Seppo J. Ovaska,et al. A modified Elman neural network model with application to dynamical systems identification , 1996, 1996 IEEE International Conference on Systems, Man and Cybernetics. Information Intelligence and Systems (Cat. No.96CH35929).

[4] Jeffrey L. Elman,et al. Finding Structure in Time , 1990, Cogn. Sci..

[5] Duc Truong Pham,et al. Training of Elman networks and dynamic system modelling , 1996, Int. J. Syst. Sci..

[6] D. T. Psent,et al. Training of Elman networks and dynamic system modelling , .

[7] 李宁,et al. Optimizing neural network forecast by immune algorithm , 2006 .

[8] DeLiang Wang,et al. On Temporal Generalization of Simple Recurrent Networks , 1996, Neural Networks.

[9] Mohamed Ibnkahla,et al. Nonlinear System Identification Using Neural Networks Trained with Natural Gradient Descent , 2003, EURASIP J. Adv. Signal Process..