Wind Speed Prediction Using Wavelet Transform and GA Trained Artificial Neural Network

Dr. Sanju Saini and Muskaan Ahuja

Wind power generation is one of the fastest increasing energy conversion systems since last few decades (generally due to the rising concerns for global warming, financial incentives (from governments), improvement in power electronic design and manufacturing etc.). With the increasing involvement of wind power in the rising power system, a precise prediction method of wind power is very much necessary for the system operator with a purpose to use the wind production in economic development, unit commitment and problem of reserve allocation. In this paper, an intelligent prediction model (using artificial neural network (ANN)) is used for this purpose and an attempt is made to improve its performance by using an optimization technique like genetic algorithm and wavelet transform. The performance is evaluated on the basis of statistical indicator such as MAE (mean absolute error). It has been observed that ANN model trained by using genetic algorithm works better than ANN trained by using back propagation learning.

Volume 11 | 09-Special Issue

Pages: 198-204

DOI: 10.5373/JARDCS/V11/20192555