NEURAL-NETWORK ALPHA-TH-ORDER INVERSE CONTROL OF THYRISTOR CONTROLLEDSERIES COMPENSATOR

This paper focuses on using neural network alpha th-order inverse cont rol strategy (X. Dai et al., Neural network alpha th-order inverse sys tem method for the control of nonlinear continuous systems, IEE Procee dings-Control Theory and Application (in press); J. Liu, X. Dal, Feasi bility of neural network alpha th-order inverse system control method, Power Syst. Autom. 21 (5) (1997) 15-18; C. Li, Y. Feng, Inverse Syste m Method of Multi-Variable Nonlinear Control, Tsinghua University Pres s, China, 1991) in the control of thyristor controlled series compensa tor (TCSC) of transmission system. The identification of neural networ k alpha th-order inverse system and the design of additional linear co ntroller are studied for the cases of a single-machine infinite bus sy stem as well as a single-machine infinite bus system simplified from a n actual power system. The action of this new strategy on power system stability is also studied. Simulation results show comparing with con ventional PID control strategy, the control strategy proposed in this paper is strongly adaptive and the stability of the power systems can be greatly enhanced. (C) 1998 Elsevier Science S.A. All rights reserve d.