CONCEPTUAL DEVELOPMENT OF OPTIMAL LOAD FREQUENCY CONTROL USING ARTIFICIAL NEURAL NETWORKS AND FUZZY SET-THEORY

Design of a new adaptive optimal load frequency control (LFC) system u sing an artificial neural network (ANN) and its performance irt a comp uter simulation of the two-area load frequency control problem is demo nstrated In order to achieve minimization of the excursions of area co ntrol errors (ACEs), their integrals and control vectors about the ste ady stare values over a wide range of operating conditions, it is desi rable to adapt elements of the feedback matrix K. The control system d esign is based on the pattern recognition principle and ill implementa tion on the parallel distributed computational architecture of artific ial neural networks. A comparison between neural-net based LFC and fuz zy logic based LFC is presented. The fuzzy logic controller based on a set of fuzzy logic operations that are performed on controller inputs , provides a means of converting a linguistic control strategy based o n expert knowledge into an efficient control strategy. Digital simulat ions of two-area power system dynamic response subject to a disturbanc e of a step load change under different oper ating conditions are perf ormed and the comparisons of conventional integral control, stare spac e optimal control and neural-net based optimal control are presented, The effectiveness of the proposed fuzzy logic controller is illustrate d through the comparison of dynamic performances obtained using fuzzy controller and the state-space optimal controller.