Add a New Input to Neural Network with Genetic Learning Algorithm to Improve Short-Term Load Forecasting

Short-term load forecasting (STLF) plays an essential role in the economic system and save the country's electricity supply. In this paper, are used a neural network with genetic learning algorithm for forecasting the electric power load of Khorasan area in Iran. Because the importance of neural network inputs, select the optimal inputs is deducted errors system. Consumption load is a nonlinear function of various factors such as weather conditions and periodic changes. This paper proposed a new variable together with the data load and temperature parameters for the problem of STLF. The variable obtained from the load curves and effect of periodic changes. The obtained results indicate that the proposed variable is effective for forecasting the short term load in electric power systems.