Solar Photovoltaic Power Forecasting

Table 1

Summary of the literature review in the very short-term time horizon.

References	Methods	Inputs	Best results
[12]	Cloud speed forecast (VOF and CCM forecasting techniques)	PNG images	FS = 0.19
[13]	NWP model, sky images, satellite images, cloud cover, and the time series models	—	—
[14]	SVR-2D	Past PV power and weather data	MRE = 9.65%; MAID = 108.33 kW; ICP = 73.07%
[15]	Cloud speed persistence	Solar power output data of 96 inverters and cloud motion data	RMSE = 4%
[16]	Machine learning techniques based on ANNs and support vector regression (SVR)	Past data of PV power and weather parameters	—
[17]	Regression tree (RT) method applied for 3 cases (cloudy day, clear day, and yearlong)	Past data of weather parameters and PV power	NRMSE = 13.8 %