Estimation Prediction Interval of Solar Irradiance Based on Just-in-Time Modeling for Photovoltaic Output Prediction

Tomohide Yamazaki, Hayato Homma, Shinji Wakao, Yu Fujimoto, Yasuhiro Hayashi

Research output: Contribution to journalArticlepeer-review

8 Citations (Scopus)

Abstract

Photovoltaic (PV) systems have recently attracted considerable attention in the context of environmental problems, antinuclear power movements, and energy problems. Therefore, the large-scale introduction of PV systems is expected in the near future. But the connection of many PV systems to the power system leads to problems. For example, the system voltage often drifts from the norm when reverse power flows increase. Accordingly, optimal system operation is required in order to make maximum use of solar energy to the maximum by installing energy buffers such as storage batteries. In particular, in forecast information, knowledge of the reliability as well as the predicted solar irradiance is essential for effective operation. In this paper, we propose a way of estimating the prediction interval of solar irradiance as an index of reliability by using just-in-time modeling. We consider the accuracy of the prediction interval under many conditions and derive a high-precision estimation method. In addition, we also discuss about the future subjects of study.

Original languageEnglish
Pages (from-to)1-10
Number of pages10
JournalElectrical Engineering in Japan (English translation of Denki Gakkai Ronbunshi)
Volume195
Issue number3
DOIs
Publication statusPublished - 2016 May 1

Keywords

  • forecast of solar irradiance
  • just-in-time modeling
  • kernel density estimation
  • photovoltaics
  • prediction interval

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Energy Engineering and Power Technology

Fingerprint

Dive into the research topics of 'Estimation Prediction Interval of Solar Irradiance Based on Just-in-Time Modeling for Photovoltaic Output Prediction'. Together they form a unique fingerprint.

Cite this