A new optimal prediction technique for energy demand based on CNN and improved water strider algorithm: a study on socio-economic-climatic parameters

Abstract

Rising demand and the potential decline in renewables resources have made energy production one of the most important challenges of the future. In hydropower plants, various factors can be the reasons to limit resources and energy supply. In this study, population and GDP, and climatic parameters (temperature and precipitation) are analyzed for the prediction of the energy demand. The innovation of this research is the Introduction of a developed algorithm called the Improved Water Strider Algorithm (IWSA). The results displayed that the improved algorithm has the faster convergence and the lowest error with the value of 1.55, 1.48, 1.44, and 0.78 in population, and GDF, temperature, and precipitation, respectively. This model has the highest correlation coefficient with the value of 0.77, 0.79, 0.8, and 80 in population, GDF, temperature, and precipitation, respectively. Also, this model with the highest correlation of 0.91, and the lowest error of 0.47 can have the best performance. Therefore, after confirming this proposed method, energy demand is predicted. The results showed that among the four input parameters of the CNN-IWSA model the precipitation parameter can have a significant effect on limiting resources and increasing demand because it can directly affect the amount and timing of energy production distribution. The results also show an increasing trend in energy demand for the next 20 years. These results can be of great help to hydrologists and energy managers in controlling and supplying energy.