引用本文: | 王守相,王亚旻,刘 岩,张 娜.基于经验模态分解和ELM神经网络的逐时太阳能辐照量预测[J].电力自动化设备,2014,34(8): |
| WANG Shouxiang,WANG Yamin,LIU Yan,ZHANG Na.Hourly solar radiation forecasting based on EMD and ELM neural network[J].Electric Power Automation Equipment,2014,34(8): |
|
摘要: |
准确的太阳能辐照量预测对于光伏发电系统具有重要意义。提出一种基于经验模态分解(EMD)和ELM神经网络的逐时辐照量组合预测模型。首先,根据预测日的环境信息,构建相似日逐时辐照量时间序列;然后,将时间序列进行EMD,分解为具有不同频率的信号,并对每个信号建立ELM神经网络预测模型;最后,将不同信号的预测值相加便可得到原始辐照量序列的预测值。算例比较表明,所提方法比传统的预测方法具有更高的预测准确度和更快的运算速度。 |
关键词: 经验模态分解 ELM神经网络 太阳能 辐照量 预测 模型 |
DOI: |
分类号: |
基金项目:基金项目:国家自然科学基金资助项目(51077098);国家高技术研究发展计划(863计划)项目(2011AA05A107);国家电网公司科技项目(ZDK/GW021-2012) |
|
Hourly solar radiation forecasting based on EMD and ELM neural network |
WANG Shouxiang1, WANG Yamin1, LIU Yan2, ZHANG Na1
|
1.Key Laboratory of Smart Grid of Ministry of Education,Tianjin University,Tianjin 300072,China;2.China Electric Power Research Institute(State Grid Metering Center),Beijing 100192,China
|
Abstract: |
Accurate solar radiation forecasting is essential for photovoltaic generation system. An hourly solar radiation forecasting model based on EMD(Empirical Mode Decomposition) and ELM neural network is proposed. The hourly solar radiation sequence of similar days is built according to the environmental information of the forecast day,which is then decomposed to signals with different frequencies by EMD. An ELM neural network forecasting model is built for each signal. The forecasting value of original solar radiation sequence is obtained by adding up the forecasting values of different signals. Case study shows that,compared with the traditional forecasting methods,the proposed method has higher forecasting accuracy and faster computation speed. |
Key words: EMD ELM neural network solar energy radiation forecasting models |