OELF: short term load forecasting for an optimal electrical load forecasting using hybrid whale optimization based convolutional neural network

Over the past few decades, models have been developed to accurately predict electrical charges. Long-term electricity forecast is the expansion of electrical equipment company management in the future. The short-term forecast for fuel and unit maintenance provides the information needed to systematically manage the unit's day-to-day operations and commitments. In this paper, we present the best in-class estimation method (OELF) to overcome micro grid problems. The proposed OELF system uses a hybrid convolutional neural network (CNN) and improved whale (IWO) to meet demand and facilitate
economic growth. The main purpose of the CNN-IWO algorithm is to calculate the maximum demand for the micro grid
and optimize the controllable load capacity for each project test. By investing in materials that reduce the performance of the micro grid, we can adjust the load on the micro grid and increase the controllable load. Therefore, OELF system for expanding micro grid expansion must carefully design cost control and load control strategies. The result showed that the performance of proposed OELF system is very effective in terms of mean MAPE and mean RMSE. The results clearly shows the average mean MAPE of proposed OELF system is 7.24%, 6.02% and 8.27% lower than the existing fuzzy based system in terms of 2 days, 1 days and 1 h ahead precision. The average mean RMSE of proposed OELF system is 9.37%, 8.34% and 5.41% lower than the existing fuzzy based system in terms of 2 days, 1 days and 1 h ahead precision.