Self-learning control of nonlinear system using Back-propagation neural networks.

Back-Propagation 신경 회로망을 이용한 비선형 시스템의 자기 학습 제어

A new algorithm is proposed to identify the structure and the parameters of the nonlinear discrete-time plant with only the unknown dynamics and the weak informations about its structure. The proposed algorithm is constructed with the compensation method of weghing values using its previous derivatives and with the efficient technique updating self-learning coefficients. The result in this application is thought to prove the effectiveness of the algorithm proposed in this paper and its superiority to the conventional ones.