Jing Wang and Hui Chen, Identification of Nonlinear Dynamic Systems Using Diagonal Recurrent Neural Networks, J. Univ. Sci. Technol. Beijing, 6(1999), No. 2, pp. 149-151.
Cite this article as:
Jing Wang and Hui Chen, Identification of Nonlinear Dynamic Systems Using Diagonal Recurrent Neural Networks, J. Univ. Sci. Technol. Beijing, 6(1999), No. 2, pp. 149-151.
Jing Wang and Hui Chen, Identification of Nonlinear Dynamic Systems Using Diagonal Recurrent Neural Networks, J. Univ. Sci. Technol. Beijing, 6(1999), No. 2, pp. 149-151.
Citation:
Jing Wang and Hui Chen, Identification of Nonlinear Dynamic Systems Using Diagonal Recurrent Neural Networks, J. Univ. Sci. Technol. Beijing, 6(1999), No. 2, pp. 149-151.
Information Engmeering School, University of Science and Techaology Beijing, Beijing 100083, China
中文摘要
In order to apply a new dynamic neural network-Diagonal Recurrent Neural NetWork (DRNN) to the system identification of nonlinear dynamic Systems and construct more accurate system models, the structure and learning method (DBP algorithm) of the DRNN are Presented. Nonlinear system characteristics can be identified by presenting a set of input / output patterns to the DRNN and adjusting its weights with the DBP algorithm. Experimental results show that the DRNN has good performances in the identification of nonlinear dynamic systems in comparison with BP networks.
In order to apply a new dynamic neural network-Diagonal Recurrent Neural NetWork (DRNN) to the system identification of nonlinear dynamic Systems and construct more accurate system models, the structure and learning method (DBP algorithm) of the DRNN are Presented. Nonlinear system characteristics can be identified by presenting a set of input / output patterns to the DRNN and adjusting its weights with the DBP algorithm. Experimental results show that the DRNN has good performances in the identification of nonlinear dynamic systems in comparison with BP networks.
下载: