Automatic generator re-dispatch for a dynamic power system by using an artificial neural network topology

Abstract

In recent years, the number of severe fault situations and blackouts worldwide has increased with the growth of large interconnected power system networks. This paper attempts to investigate the effect of a contingency (N-1) on rotor angle stability and thermal line flow for a dynamic power system. In addition, a solution is presented to eliminate system instability by providing an automatic generator re-dispatch instantly after a disturbance. Based on the ability of an Artificial Neural Network (ANN), it is possible to model a mathematical relationship between a power system disturbance and a control action due to the fast response of an ANN system. This relationship is obtained by the neurons between the input and output layers of the ANN topology. The completed model and data knowledge preparation process were successfully tested on an IEEE 9-bus test system. The ANN was able to provide a control action in a very short time period with high accuracy. An optimal amount of generator re-dispatch in Megawatt (MW) can contribute towards eliminating bus voltage and thermal line flow violations or unstable power system operation.