Assessment of hydroelectric potential under climate change and hydrological parameters based on soft-computing: A case study

Abstract

Electricity is recognized as one of the most pervasive government services. Therefore, accurate forecasting of electricity demand has a great impact on electricity management. In this paper, the monthly electricity demand has been forecasted. For this objective, an integrated Artificial Neural Network (ANN) with a developed metaheuristic algorithm has been used to forecast short-term electricity demand. The improved algorithm reduces errors in the optimization process by preventing a fall into local optimization. This algorithm is called the Developed African Vulture Optimization (DAVOA) algorithm. The finding indicates the annual contribution of temperature, relative moisture, wind speed, and cloudiness to forecast short-term electricity demand by about 70%, 12%, 10%, and 8%, respectively. Therefore, temperature, with a correlation of 0.70%, is an important variable that affects short-term energy demand. Moreover, the finding of short-term energy demand based on the optimal ANN method indicates that the electricity demand change in January, February, July, August, and September is greater than in other months because the temperature changes more drastically in these months. Therefore, the use of heating and cooling tools will increase compared to other months, and consequently, the energy demand for the energy supply of cooling and heating tools will increase.