Adaptive control for multi-machine power systems using genetic algorithm and neural network

This paper presents an adaptive control technique for the variable series capacitor (VSrC) using a recurrent neural network (RNN). Since parameters of the controller determined by genetic algorithm (GA), which is one of the optimization algorithms, are optimum for only one operating point, it is possible not to realize good control performance against variations of the operating point and fault point. Then, the adaptive controller proposed in this paper consists of the optimum controller using GA and the recurrent neural network (RNN). As the RNN is learned on-line, robust control performance can be realized in various conditions. The effectiveness of this control method is verified by simulation results of a multi-machine power system.