The evaluation of solely renewable energy based virtual power plants potential for ancillary services provision

Abstract

High penetration of variable renewable energy sources in the utility grid drives the increasing need for ancillary services (AS) in electricity markets. Wind and Photovoltaics power plants are proven technically and economically capable of providing AS to system operators. To estimate the potential of AS provision by solely renewables based Virtual Power Plant (VPP), an evaluation methodology from the perspective of machine learning (ML) is proposed in this paper. Then, the methodology is tested using open-source datasets that contain hourly power output from wind and solar power plants located in 21 regions of France.