Study of wind speed prediction method for wind power generation
Project/Area Number |
26340107
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Research Field |
Design and evaluation of sustainable and environmental conscious system
|
Research Institution | Hiroshima Institute of Technology |
Principal Investigator |
Maeda Shunji 広島工業大学, 工学部, 教授 (00626799)
|
Project Period (FY) |
2014-04-01 – 2017-03-31
|
Project Status |
Completed (Fiscal Year 2016)
|
Budget Amount *help |
¥4,680,000 (Direct Cost: ¥3,600,000、Indirect Cost: ¥1,080,000)
Fiscal Year 2016: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2015: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2014: ¥2,210,000 (Direct Cost: ¥1,700,000、Indirect Cost: ¥510,000)
|
Keywords | 風速予測 / ニューラルネットワーク / ベクトル自己回帰モデル / サポートベクトル回帰 / 重回帰分析 / 風力発電 / 因果関係 / インパルス応答関数 / 類似性 / 自己回帰モデル / 部分空間法 |
Outline of Final Research Achievements |
Nobel method of forecasting the wind speed for short time is proposed. In wind speed prediction using a neural network, the model is generated every time using past similar wind speed data. To select the past wind speed data similar to the present situation, the coefficients of the vector auto regressive (VAR) model are focused. The VAR coefficients are used for the selection of the data and the past wind speed data which have high VAR coefficients are input to the neural network. The proposed method is confirmed that combination of the above idea is effective. Specifically, the wind speed prediction error (average absolute value error) , one hour ahead in the Tohoku region is 1m/s that is close to the target.
|
Report
(4 results)
Research Products
(18 results)