| 引用本文: | 张 露,卢继平,梅亦蕾,朱三立.基于不同优化准则的风电功率预测[J].电力自动化设备,2015,35(5): |
| ZHANG Lu,LU Jiping,MEI Yilei,ZHU Sanli.Wind power forecasting based on different optimization criterions[J].Electric Power Automation Equipment,2015,35(5): |
|
| 摘要: |
| 为了获得较好的风电功率预测综合评价指标和提高风电功率预测精度，提出基于不同优化准则的风电功率预测模型。首先，通过基于贴近度的单项预测模型的择优方法得到较高精度的单项预测模型，然后利用选择出的单项预测模型分别以平均相对误差最小、平均绝对误差最小和均方根误差最小为优化准则建立不同的组合预测模型，最后利用灰色关联度分析方法确定每种组合预测模型在优化模型中的权系数，进而得到优化模型。以风电场的实际数据进行验证，结果表明：与各单项预测模型、不同优化准则的组合预测模型及其他组合模型相比，所提优化模型的整体误差指标较小，有效地提高了预测精度，证明了所提模型的有效性和实用性。 |
| 关键词: 风电 贴近度 单项预测模型 优化准则 组合预测模型 灰色关联度分析 优化 模型 |
| DOI： |
| 分类号: |
| 基金项目:输配电装备及系统安全与新技术国家重点实验室自主研究项目（2007DA10512712205） |
|
| Wind power forecasting based on different optimization criterions |
|
ZHANG Lu, LU Jiping, MEI Yilei, ZHU Sanli
|
|
State Key Laboratory of Power Transmission Equipment & System Security and New Technology，Chongqing University，Chongqing 400044，China
|
| Abstract: |
| In order to obtain better comprehensive evaluation index and improve the accuracy of wind power forecasting，a wind power forecasting model based on different optimization criteria is proposed． The approach degree is adopted to select the single forecasting models with higher accuracy，based on which，different combination forecasting models are built with the minimum average relative error，minimum mean absolute error or minimum root mean square error as the optimization criterion respectively. The grey correlation analysis method is applied to determine the weight coefficients of each combination model and then obtain the optimized model. The verification with the actual data of a wind farm show that，compared with single forecasting models，combination forecasting models based on different optimization criteria and other combination models，the proposed optimized model has smaller overall error and improves the forecasting accuracy effectively，which proves its effectiveness and practicability. |
| Key words: wind power approach degree single forecasting model optimization criterion combination forecasting model grey correlation analysis optimization models |
