Power System Frequency Estimation Using Neural Network and Genetic Algorithm
Monika Gupta, Smriti Srivastava, J. R. P. Gupta · 2008
Frequency is an important parameter in power system monitoring, control and protection. This paper shows how a combination of neural networks and genetic algorithm can be used to estimate power system frequency. Neural networks on the other hand offer great advantages in learning, adaptation, fault tolerance and parallelism. Genetic algorithm is a parallel global search technique that emulates natural genetic operators. In the proposed algorithm learning of weights of neural networks is done using genetic algorithm. The results obtained by simulation show better performance of the proposed control structure when compared with traditional error back propagation and least mean square algorithm. The performance of the algorithm is studied through simulations at different situations of power system.