The growing resource shortage and environmental concerns have forced mankind to develop and utilize renewable energy sources. The penetration of solar photovoltaic (PV) power in the electricity market has been increasing over the past few decades due to its low construction costs, zero pollution nature, and enormous support from governments. However, the intermittency and randomness of PV power also cause huge grid fluctuations which limit its integration in the system. An accurate forecasting of solar PV power generation and optimization of operation and maintenance (O&M) management are essential for further development of the solar PV farms.
A great number of related researches have been done in recent years. A review of PV power generation forecasting techniques together with their brief applications on the optimization of O&M management is presented in this paper. Machine learning methods are thought to be the most suitable at the present stage because of their ease of implementation and their capability in processing non-linear, complex data sets. Typical forecasting accuracy measures are summarized and further applications of PV power forecasting on the O&M management are also presented.