吴小涛1,袁晓辉2,袁艳斌3,张东寅4.基于混合变分模态分解模型的短期风速预测[J].水电能源科学,2019,37(1):195-198
基于混合变分模态分解模型的短期风速预测
Short-term Wind Speed Prediction Based on Mixed Variational Model Decomposition Model
  
DOI:
中文关键词:  风速预测  最优变分模态分解  蝙蝠算法  最小二乘支持向量机
英文关键词:wind speed prediction  optimal variational mode decomposition  bat algorithm  least squares support vector machine
基金项目:国家自然科学基金项目(41571514);黄冈师范学院博士基金项目(201828603);中央高校基本科研业务费专项资金资助项目(2017KFYXJJ204)
作者单位
吴小涛1,袁晓辉2,袁艳斌3,张东寅4 1. 黄冈师范学院 数学与统计学院 湖北 黄冈 438000 2. 华中科技大学 水电与数字化工程学院 湖北 武汉 430074 3. 武汉理工大学 资源与环境学院 湖北 武汉 430070 4. 国网湖北省电力有限公司 经济技术研究院 湖北 武汉 430077 
摘要点击次数: 563
全文下载次数: 
中文摘要:
      针对风速时间序列不稳定导致其难以准确预测的问题,提出一种基于最优变分模态分解(OVMD)和蝙蝠算法(BA)优化最小二乘支持向量机(LSSVM)的短期风速预测模型。采用OVMD技术,将原始风速时间序列先分解为若干个相对稳定的分量序列,然后对各个分量分别建立LSSVM模型进行预测,并采用蝙蝠算法优化LSSVM中的参数,最后对优化的分量预测模型的预测值求和,即得到原始风速序列的预测值。算例分析表明,该模型具有较高的预测精度,能有效跟踪风速的变化规律。研究成果为短期风速预测提供了新思路。
英文摘要:
      Aiming at the problem that wind speed time series are always instability so that it is difficult to predict accurately,this paper proposed a combined model based on optimal variational mode decomposition ( OVMD) and the least squares support vector machine (LSSVM) optimized by bat algorithm (BA) for short-term wind speed prediction.Using OVMD technology, the original wind speed time series was decomposed into several relatively stable sequences. Then the LSSVM model was established to predict each component, and the bat algorithm was used to optimize the parameters of LSSVM. Finally, the prediction value of each component prediction model was accumulated to obtain the prediction value of the original wind speed series. The experimental results show that the model has higher prediction accuracy, and can effectively track the change law of wind speed. The research results provide new idea for short-term wind speed prediction.
查看全文  查看/发表评论  下载PDF阅读器