Short-term Prediction of Photovoltaic Generation Power based on Chaotic Behaviors and Power Smoothig Control Methods

2015 Fiscal Year Final Research Report

• Project/Area Number: 24560333
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Power engineering/Power conversion/Electric machinery
• Research Institution: Okayama University
• Principal Investigator: Funabiki Shigeyuki 岡山大学, 自然科学研究科, 教授 (60108123)
• Co-Investigator(Kenkyū-buntansha): TAKAHASHI Akiko 岡山大学, 大学院自然科学研究科, 助教 (10644213)
• Project Period (FY): 2012-04-01 – 2016-03-31
• Keywords: 太陽光発電 / 発電電力変動 / カオス / 予測 / 電力平滑化 / エネルギー貯蔵装置

Outline of Final Research Achievements

We developed a short-term prediction method for electric power generated by the photovoltaic generation system using the Chaos Theory (Chaos short-term prediction). The experimental verification showed that the proposed prediction method can predict generation power with error of 20 %. In the subsequent step, we proposed a novel method to predict generation power using a series of difference vectors for improving the time delay in the chaos short-term prediction. It is found from the comparative study of the power smoothing control methods that the exponential smoothing is superior to the moving average from a viewpoint of the reduction in the storage capacity of energy storage device.

Free Research Field

工学