Evaluation of the adaptive behaviour of an artificial neural network controller for a hybrid power compensator

Abstract

AbstractAn artificial neural network (ANN) is used to optimally control a hybrid power compensator (HPC) consisting of a static VAR compensator (SVC) and a dynamic compensator (DC) during the compensation of nonlinear loads. The HPC must be controlled to meet minimum requirements in terms of power factor and harmonic distortion, while at the same time minimising its total cost. The ability of the ANN to adapt its output when subjected to a dynamic power system environment to achieve the above, is evaluated. The power system dynamics in this paper constitute line impedance changes and changes in the load conditions of other consumers coupled to the same network. A state space model of the power distribution network together with the HPC forms the basis of evaluation of the mentioned controller. The results obtained reveal that the ANN can meaningfully adapt its output to optimise the HPC performance.