Enhanced Recurrent Extreme Learning Machine Using Gray Wolf Optimization for Load Forecasting

Ever increasing demand of electricity necessitates cheaper electric supply. Electricity forecasting plays very important role in order to reduce operational cost of generation, transmission and delivery system. In this paper, a novel electricity forecasting technique enhanced recurrent extreme learning machine (ERELM) has been proposed. It forecasts load using recurrent extreme learning machine (RELM) and optimizes weights and biases using gray wolf optimizer (GWO). It is back propagation network that gives context neurons to input and hidden layers after adjusting weights as per root mean square (RMSE). Cross validation of obtained results is performed using monte carlo as well as kfold methodology. Accuracy results are discussed in detail using mean absolute error (MAE), mean square error (MSE) and root mean square error (RMSE). Results show that proposed technique is better than previous techniques and can be used for modeling effective dynamic systems.