Page 49 - 《中国电力》2026年第5期

张怀天等：基于 Transformer-集成学习的配电网短期负荷预测方法 2026 年第 5 期

Construction, 2024, 45(12): 162–173. LIANG Rui, JIN Mohan, ZHU Huijun, et al. Net load interval
[23] KONG W C, DONG Z Y, JIA Y W, et al. Short-term residential load forecasting in rural areas based on multi-scenario clustering and
forecasting based on LSTM recurrent neural network[J]. IEEE explainable temporal fusion transformers network[J]. Power System
Transactions on Smart Grid, 2019, 10(1): 841–851. Technology, 2025, 49(12): 5019–5027, I0016–I0018.
[24] 杨 国 华 , 祁 鑫 , 贾 睿 , 等 . 基 于 CEEMD-SE 的 CNN & LSTM- [30] WANG C, WANG Y, DING Z T, et al. A transformer-based method
GRU 短期风电功率预测 [J]. 中国电力, 2024, 57(2): 55–61. of multienergy load forecasting in integrated energy system[J]. IEEE
YANG Guohua, QI Xin, JIA Rui, et al. Short-term wind power Transactions on Smart Grid, 2022, 13(4): 2703–2714.
forecast based on CNN & LSTM-GRU model integrated with [31] ZHAO H S, WU Y C, MA L B, et al. Spatial and temporal attention-
CEEMD-SE algorithm[J]. Electric Power, 2024, 57(2): 55–61. enabled transformer network for multivariate short-term residential
[25] 茹瑶, 赵永宁, 叶林, 等. 超短期 LSTM 风电功率预测模型的混合 load forecasting[J]. IEEE Transactions on Instrumentation and
专家模块化代理解释方法 [J]. 电力建设, 2024, 45(11): 114–124. Measurement, 2023, 72: 2524611.
RU Yao, ZHAO Yongning, YE Lin, et al. Modular surrogate [32] 俞 胜 , 孙 可 , 蔡 华 , 等 . 结 合 极 端 梯 度 提 升 决 策 树 与 改 进
interpretation method based on decision tree mixture of experts for Informer 的短期电力负荷预测方法 [J]. 中国电力, 2025, 58(10):
ultra-short-term LSTM wind power forecasting model[J]. Electric
195–205.
Power Construction, 2024, 45(11): 114–124.
YU Sheng, SUN Ke, CAI Hua, et al. A short-term power load
[26] 黄文琦, 梁凌宇, 王鑫, 等. 基于变量选择与 Transformer 模型的中
forecasting method combining extreme gradient boosting decision
长期电力负荷预测方法 [J]. 浙江大学学报 (理学版), 2024, 51(4):
tree with an improved informer[J]. Electric Power, 2025, 58(10):
483–491, 500.
195–205.
HUANG Wenqi, LIANG Lingyu, WANG Xin, et al. Mid-long term
[33] 孟浩, 徐飞, 符帅, 等. 考虑温控型负荷特性影响的集群用户超短期
power load forecasting based variable selection and transformer
负荷预测方法 [J]. 中国电力, 2025, 58(12): 63–72, 85.
model[J]. Journal of Zhejiang University (Science Edition), 2024,
MENG Hao, XU Fei, FU Shuai, et al. Ultra-short-term load
51(4): 483–491, 500.
forecasting method for aggregated users considering the impact of
[27] 孟衡, 张涛, 王金, 等. 基于多尺度时空图卷积网络与 Trans-
temperature-controlled load characteristics[J]. Electric Power, 2025,
former 融合的多节点短期电力负荷预测方法 [J]. 电网技术, 2024,
58(12): 63–72, 85.
48(10): 4297–4305.
[34] Srivastava N, Hinton G, Krizhevsky A, et al. Dropout: a simple way
MENG Heng, ZHANG Tao, WANG Jin, et al. Multi-node short-term
to prevent neural networks from overfitting[J]. The journal of
power load forecasting method based on multi-scale spatiotemporal
machine learning research, 2014, 15(1): 1929–1958.
graph convolution network and transformer[J]. Power System
[35] Cheng B, Chen Y. Open Datasets for grid modeling and
Technology, 2024, 48(10): 4297–4305.
visualization: an Alberta power network case[J]. arXiv preprint
[28] 范士雄, 李东琦, 郭剑波, 等. 基于时变滤波经验模态分解-重构和
arXiv: 2504, 07870: 2025.
独立自注意力机制的 iTransformer 超短期负荷预测方法 [J]. 电网

技术, 2025, 49(6): 2436–2445. 作者简介：
FAN Shixiong, LI Dongqi, GUO Jianbo, et al. Ultra-short-term load 张怀天（1994），女，硕士，工
forecasting method based on time-varying filter empirical mode 程师，从事配电网运行控制研究，E-mail：
decomposition-reconstruction and iTransformer with stand-alone huaitianz@163.com。
self-attention mechanism[J]. Power System Technology, 2025, 49(6):
2436–2445.
[29] 梁睿, 金沫含, 朱慧君, 等. 基于多场景聚类和可解释时间融合
Transformer 网络的乡村地区净负荷区间预测 [J]. 电网技术, 2025,
（责任编辑 王文诗）
49(12): 5019–5027, I0016–I0018.

45