Research on the Application of Machine Learning in Photovoltaic Power Station

With the increasingly prominent environmental problems, clean energy is gradually replacing fossil energy and changing the existing energy structure is the only way to achieve sustainable development in the future. This paper mainly studies the application of machine learning in photovoltaic power stations from three aspects of photovoltaic forecasting, data monitoring and fault detection, and aims to explore the development direction of machine learning in the next stage. Though study these parts, the paper discusses the current monitoring methods of photovoltaic forecasting data, and comes to the development path of in-depth data mining, multi-source data acquisition and multi-system integration. It is concluded that the comprehensive application of multiple prediction methods and monitoring methods will be the key to improve the accuracy of prediction in the future, including the intelligent detection method, unintelligent detection method and UAV (unmanned aerial vehicle) monitor method.