A Recent Invasion Wave Of Deep Learning In Solar Power Forecasting Techniques Using Ann

Since 2000, solar power has grown rapidly to meet the electricity demand for daily life, industry, agriculture, service… In the development of solar energy, forecasting the generation capacity of solar power plants plays an important role in stable operation, planning, and balancing their electricity production. Many models and techniques have been developed for shorterm, medium, and long term solar generation forecasting. Research literature from 2000 to the present shows that Artificial Neural Network (ANN)-based predictive models perform better than other models due to their ability to solve complex nonlinear problems Their accuracy can be further improved by combining the two models in forecasting or processing input data. This paper compares the accuracy, popularity, and forecasting scope of solar power generation forecasting methods using the ANN model in combination with other forecasting models. Currently, the combined forecasting models of Long Short-Term Memory (LSTM) and Gate Recurrent Unit (GRU) with other forecasting methods are being strongly developed, playing an important role in predicting the generation capacity of solar power plants.