The forecasting of solar energy based on Machine Learning

Recently, the majority of countries are seeking to redefine their energy mix, due to the global warming and climate change concerns. As a result, the world is moving towards the use of renewable energy to produce electricity. In this sense, solar energy has become one of the main strengths of the most promising renewable resources for mass power generation. Solar energy forecasting is a very active field of research. Its objective is to reduce the uncertainty of variable renewable production. Statistical, physical, machine learning and hybrid approaches were used. However, research contributions in this area remain immature. Therefore, the search for more relevant methods, adapted to this field, will be important because it would improve accuracy, in order to provide the world with many economic and environmental benefits. In this context, this article discusses the fundamental principles of the machine learning based approaches.