INTELLIGENT CURRENT CONTROLLER FOR AN HVDC TRANSMISSION LINK

This paper describes an intelligent current controller for the fast an d flexible control of an HVDC trans mission link using Artificial Neur al Network (ANN) and Fuzzy Logic (FL) paradigms. A simple yet effectiv e ANN architecture is present-ed with am-line adaptation of the activa tion function and learning parameters. Two methods of adapting the lea rning parameters are presented. In the first method, a heuristic appro ach to evaluate the learning rate as a polynomial of an energy functio n is considered. In the second method, a Fk based on-line adaption of the learning parameters is discussed. Performance of ANN, ANN-FL based and PI controllers are compared. A feasibility analysis is carried ou t to implement ease proposed neural controller algorithm in real-time.