One of the most significant renewable energy sources for the production of sustainable power is solar energy. However, due to meteorological and environmental conditions including temperature, humidity, cloud cover, visibility, and sun radiation, the amount of electricity generated by solar systems fluctuates greatly. Stable integration of renewable energy into power grids and effective energy management depend on accurate solar power generation predictions. The machine learning-based method for predicting solar power generation utilising sophisticated data processing techniques is presented in this paper. The suggested method makes use of a dataset that includes temporal characteristics like hour, day, month, and year in addition to other environmental aspects. To increase the dataset's quality and boost model performance, data preprocessing techniques including feature extraction and normalisation are used. The analysed data is used to train machine learning algorithms that forecast solar power generation. Through an interactive interface, users may view datasets, train models, and carry out predictions with this web-based tool. Standard metrics like R2 Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE) are used to assess the performance of the suggested model. The results of the experiments show that the suggested method may accurately predict solar power generation. The proposed technology can help renewable energy providers and energy planners make better decisions and use solar electricity.
Mahesh et al. (Sun,) studied this question.