화학공학소재연구정보센터
IEE Proceedings-Control Theory & Applications, Vol.141, No.5, 305-314, 1994
Block Partial Derivative and Its Application to Neural-Net-Based Direct-Model-Reference Adaptive-Control
The concept of block partial derivative and the associated algebra are introduced. The algebra is then applied to direct-model-reference adaptive control (MRAC) of discrete-time nonlinear plants. Three MRAC algorithms are developed employing identification of the forward model, extended MIT rule and extended SPR rule. Local-convergence properties and results of simulation study are presented.