Page 7 - 电力与能源2023年第五期
P. 7
第 44 卷 第 5 期 电力与能源
2023 年 10 月 429
DOI:10.11973/dlyny202305001
研究与探索
基于 LSTM 组合模型的短期电力负荷预测
李 盖 ,林余杰 ,吴成坚 ,徐文进 2
1
1
1
(1. 国网浙江省电力有限公司乐清市供电公司,浙江 乐清 325600;2. 国网上海市电力公司浦东供电公司,上海 200120)
摘 要:针对电力负荷时序难以进行精度预测的难题,提出了一种基于自适应白噪声的完备集合经验模态分
解(CEEMDAN)的 TCN-LSTM 短期电力负荷组合预测方法。首先使用 CEEMDAN 分解方法将原始负荷
序列进行分解,该方法与集合经验模态分解(EEMD)方法相比,能使序列分解更加完备,且具有更小的重构误
差。然后为了降低非平稳序列对预测精度的影响,通过平稳性检验将分解后的序列按照平稳性质分类,将非
平稳序列合并后输入 LSTM 网络预测,平稳序列则计算排列熵后重组成高排列熵的平稳序列和低排列熵的
平稳序列,再分别输入到 LSTM 网络和 TCN 网络中进行预测,最后对预测结果进行叠加得到最终的预测结
果。实证结果表明:通过按照平稳性分类和计算排列熵的方式来对 CEEMDAN 分解后的序列进行重新组合
的方法,不仅提高了模型的运算效率,同时比其他预测方法具有更高的预测精度。
关键词:自适应噪声完备集合经验模态分解;排列熵;超短期电力负荷预测
作者简介:李 盖(1988—),男,工程师,从事配电网线损管理与统计分析。
中图分类号:TM715 文献标志码:A 文章编号:2095-1256(2023)05-0429-08
Short-Term Power Load Forecasting Based on LSTM Combination Model
1
1
1
LI Gai ,LIN Yujie ,WU Chengjian ,XU Wenjin 2
(1.State Grid Yueqing City Power Supply Company,Zhejiang Electric Power Co.,Ltd.,Yueqing 325600,
Zhejiang Province,China;2.State Grid Pudmg Power Sapply Company,SMEPC,Shanghai 200120,China)
Abstract:Considering the difficulty of accurate prediction of power load time series, this research proposes a
TCN-LSTM short-term power load combination forecasting method based on complete ensemble empirical mode
decomposition with adaptive noise (CEEMDAN). First, the original load sequence is decomposed using
CEEMDAN decomposition method. Compared with ensemble empirical mode decomposition (EEMD) method,
this method can make the sequence decomposition more complete and has less reconstruction error. Then, in or⁃
der to reduce the influence of non-stationary sequences on the prediction accuracy, the decomposed sequences are
classified according to their stationary properties through stationarity test, and the non-stationary sequences are
combined and input into the LSTM network for prediction. The stationary sequences are then reconstructed into
stationary sequences with high and low permutation entropy by calculating the permutation entropy. Then they are
input into LSTM network and TCN network respectively for prediction. Finally the prediction results are superim⁃
posed to get the final prediction results. The empirical results show that the method of recombining the decom⁃
posed sequences of CEEMDAN based on stationality classification and calculation of permutation entropy not
only improves the efficiency of the model, but also has higher prediction accuracy than other prediction methods.
Key words:complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN),permutation
entropy,ultra-short-term power load forecasting
地区电力负荷的精准预测一直以来都是电力 提高经济效益和社会效益。同时负荷预测的准确
部门的重要工作,涉及电力的合理规划、调度、运 程度,也将直接影响电力系统的优化布局 [1-3] 。
营等多个方面。准确的负荷预测有助于合理地控 近年来,随着数学统计理论的不断深化以及
制电网内部发电机组的运行,保障电网运行的安 人工智能技术的不断发展,国内外众多学者运用
全与稳定,还能有效降低地区的配电和用电成本, 多种方法来对电力负荷进行预测。这些预测方法

