| 引用本文: | 李 楠,金淳熙,陶 亮,等.基于多尺度二次特征提取的短期电力负荷预测模型[J].电力系统保护与控制,2025,53(19):114-126.[点击复制] |
| LI Nan,JIN Chunxi,TAO Liang,et al.A short-term electric load forecasting model based on multi-scale secondary feature extraction[J].Power System Protection and Control,2025,53(19):114-126[点击复制] |
|
| 摘要: |
| 为充分挖掘电力负荷固有多尺度特征(multi-scale feature, MSF)间的复杂时序关系，进一步提升电力负荷预测模型性能，特别是提升对节假日负荷预测的能力，提出了一种基于多尺度二次特征提取的短期电力负荷预测模型。首先，利用Prophet算法的拟合分解功能，获取不同尺度下的负荷数据分量，并与强相关的天气数据共同构建多元数据集。其次，采用改进的特征金字塔网络(improved feature pyramid network, IFPN)匹配负荷数据的多尺度特性，并设计了卷积特征增强模块强化对节假日特征的表达能力，实现MSF的第一次提取。基于时间卷积神经网络的优势，深度挖掘一次特征间的时序依赖关系，引入SENet对特征的权重实现自适应赋值，同时完成MSF的二次提取。最后，利用鱼鹰算法优化后的Transformer模型完成负荷预测。在国内外两套负荷数据集上进行了验证，仿真结果表明所提模型的预测效果优于对比模型，特别是在对节假日负荷的预测精度上有所提高。 |
| 关键词: 短期电力负荷预测 Prophet算法 二次特征提取 改进的特征金字塔网络 多尺度时间卷积网络 |
| DOI：10.19783/j.cnki.pspc.241360 |
| 投稿时间：2024-10-14修订日期：2025-02-25 |
| 基金项目:国家自然科学基金面上项目资助(52277084) |
|
| A short-term electric load forecasting model based on multi-scale secondary feature extraction |
| LI Nan1,2,JIN Chunxi2,TAO Liang3,HUANG Liang3 |
| (1. Key Laboratory of Modern Power System Simulation Control and Green Energy Technologies, Ministry of Education,
Jilin 132012, China; 2. School of Electrical Engineering, Northeast Electric Power University, Jilin 132012, China; 3. Siping
Power Supply Company, Jilin Electric Power Company Limited of State Grid Corporation of China, Siping 136000, China) |
| Abstract: |
| To fully explore the complex temporal relationships among the inherent multi-scale features (MSF) of electrical load data and further improve the performance of electricity load forecasting models, especially their accuracy during holidays, a short-term electricity load forecasting model based on multi-scale secondary feature extraction is proposed. First, the Prophet algorithm is used to decompose and fit the load data, extracting components at different scales, which are then combined with correlated weather data to construct a multivariant dataset. Then, an improved feature pyramid network (IFPN) is employed to match the multi-scale characteristics of load data. A convolutional feature enhancement module is designed to strengthen the model’s ability to express holiday-specific features, achieving the first extraction of MSF. Leveraging the advantages of temporal convolutional neural networks, the model deeply mines the temporal dependencies among the primary features. Squeeze-and-excitation networks (SENet) is introduced to adaptively assign weights to features, completing the secondary extraction of MSF. Finally, load forecasting is performed using a Transformer model optimized by the Osprey algorithm. Validation on two domestic and international load datasets shows that the proposed model outperforms comparison models, particularly in improving prediction accuracy during holidays. |
| Key words: short term power load forecasting Prophet algorithm secondary feature extraction IFPN MSTCN |
