| 引用本文: | 范磊,卫志农,李慧杰,Kwok W Cheung,孙国强,孙永辉.基于变分模态分解和蝙蝠算法-相关向量机的短期风速区间预测[J].电力自动化设备,2017,37(1): |
| FAN Lei,WEI Zhinong,LI Huijie,Kwok W Cheung,SUN Guoqiang,SUN Yonghui.Short-term wind speed interval prediction based on VMD and BA-RVM algorithm[J].Electric Power Automation Equipment,2017,37(1): |
|
| 摘要: |
| 现有的风速预测方法大多是确定性的点预测，无法描述风速的随机性。针对该问题，建立基于变分模态分解(VMD)和蝙蝠算法-相关向量机(BA-RVM)的短期风速区间预测模型。对原始风速序列进行变分模态分解获得多个子序列；采用样本熵(SE)算法对子序列进行重组得到3类具有典型特性的分量；对各分量采用相关向量机算法分别建立预测模型。为进一步提高预测精度、缩小区间范围，引入蝙蝠算法(BA)对预测模型进行参数优化。将各分量的预测结果进行叠加求和得到一定置信水平下总体的区间预测结果。实际算例结果表明，与现有方法相比，所提区间预测方法的预测精度和区间覆盖率更高，区间宽度更窄。 |
| 关键词: 风电 风速预测 短期预测 相关向量机 变分模态分解 区间预测 |
| DOI：10.16081/j.issn.1006-6047.2017.01.015 |
| 分类号: |
| 基金项目:国家自然科学基金资助项目(51107032,61104045,51277052)；国家高技术研究发展计划(863计划)资助项目(2013AA050601) |
|
| Short-term wind speed interval prediction based on VMD and BA-RVM algorithm |
|
FAN Lei1, WEI Zhinong1, LI Huijie2, Kwok W Cheung3, SUN Guoqiang1, SUN Yonghui1
|
|
1.College of Energy and Electrical Engineering, Hohai University, Nanjing 210098, China;2.ALSTOM GRID Technology Center Co.,Ltd.,Shanghai 201114, China;3.GE Grid Solutions Inc.,Redmond 98052, USA
|
| Abstract: |
| Since the existing wind speed prediction methods are mostly of deterministic point forecasting and could not describe the randomness of wind speed, a short-term wind speed interval prediction model based on VMD(Variational Mode Decomposition) and BA-RVM(Bat Algorithm-Relevance Vector Machine) is built. VMD is used to get multiple sub-sequences from the original wind speed sequence, SE(Sample Entropy) algorithm is applied to reorganize these sub-sequences for obtaining three types of typically characteristic components, and RVM algorithm is adopted to build the forecasting model for each component. BA is introduced to optimize the model parameters for further improving the prediction accuracy and reducing the interval range. The overall interval prediction with a certain confidence level is obtained by superimposing the forecasted results of three components. Results for a practical case show that, compared with the existing methods, the proposed method can get higher forecasting accuracy, bigger interval coverage rate and smaller interval width. |
| Key words: wind power wind speed prediction short-term prediction RVM VMD interval prediction |
