Solar power generation prediction for distribution systems with high PV penetration


10th IEEE PES Innovative Smart Grid Technologies Europe, ISGT-Europe 2020, Delft, Netherlands, 26 - 28 October 2020, vol.2020-October, pp.389-393 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 2020-October
  • Doi Number: 10.1109/isgt-europe47291.2020.9248919
  • City: Delft
  • Country: Netherlands
  • Page Numbers: pp.389-393
  • Keywords: Behind-the-meter PV systems, demand prediction, distribution systems, solar power generation, NEURAL-NETWORK, LOAD, ALGORITHM
  • Middle East Technical University Affiliated: Yes


© 2020 IEEE.Distributed renewable energy sources, mainly photovoltaic (PV) systems, have become increasingly popular in recent years thanks to their environmental and economic benefits. Despite their benefits, installation of high number of PV systems brings new challenges in power system operation. Distributed PV systems at low voltage distribution systems are normally located behind the meter and hence, invisible to the system operators. This unmonitored generation may cause considerable change in the net demand from the transmission system. Uncertainty associated with the net demand may have significant impact on planning and operation of power systems. This paper aims to provide situational awareness by predicting unmonitored solar power generation in the considered distribution system. The proposed methodology predicts total solar power generation with a small number of measurements by using probabilistic knowledge on cloudiness over a city.