Page 135 - 《中国电力》2026年第5期

周专等：基于动态权重混合专家模型的超短期电力负荷预测 2026 年第 5 期

diagnosis method based on multi-pathed hierarchical mixture-of- [18] 文越, 邵必林, 曹翔, 等. 基于 Transformer-GAN 的天然气负荷数据
experts model[J]. Journal of Computer Applications, 2025, 45(1): 异常检测分析 [J]. 集成电路应用, 2025, 42(1): 58–61.
59–68. WEN Yue, SHAO Bilin, CAO Xiang, et al. Analysis of anomaly
[11] 李鹏, 罗湘淳, 孟庆伟, 等. 基于 Spearman 相关性阈值寻优和 detection of natural gas load data based on Transformer-GAN[J].
VMD-LSTM 的用户级综合能源系统超短期负荷预测 [J]. 全球能 Application of IC, 2025, 42(1): 58–61.
源互联网, 2024, 7(4): 406–420. [19] 李加文, 孙永辉, 王森, 等. 计及异常场景数据缺失的负荷超短期预
LI Peng, LUO Xiangchun, MENG Qingwei, et al. Ultra short-term 测 [J]. 电力系统自动化, 2025, 49(15): 133–143.
load forecasting of user level integrated energy system based on
LI Jiawen, SUN Yonghui, WANG Sen, et al. Ultra-short-term load
spearman threshold optimization and variational mode decomposition
forecasting considering missing data in anomalous scenarios[J].
and long short-term memory[J]. Journal of Global Energy
Automation of Electric Power Systems, 2025, 49(15): 133–143.
Interconnection, 2024, 7(4): 406–420.
[20] 曾静, 娄冰, 吕娜, 等. 基于多特征和 LOF 的用户负荷突变检测 [J].
[12] 茹瑶, 赵永宁, 叶林, 等. 超短期 LSTM 风电功率预测模型的混合
浙江电力, 2023, 42(2): 90–97.
专家模块化代理解释方法 [J]. 电力建设, 2024, 45(11): 114–124.
ZENG Jing, LOU Bing, LYU Na, et al. Abrupt user load change
RU Yao, ZHAO Yongning, YE Lin, et al. Modular surrogate
detection based on multiple features and LOF algorithm[J]. Zhejiang
interpretation method based on decision tree mixture of experts for
Electric Power, 2023, 42(2): 90–97.
ultra-short-term LSTM wind power forecasting model[J]. Electric
[21] YI C C, PENG Y, SU S, et al. Anomaly detection of photovoltaic
Power Construction, 2024, 45(11): 114–124.
power generation based on quantile regression recurrent neural
[13] 杨雪莹, 祁琪, 李启明, 等. 奖励机制与用户意愿结合的高峰期负荷
network[J]. Electric Power Systems Research, 2025, 238: 111132.
博弈调度策略 [J]. 电工技术学报, 2024, 39(16): 5060–5074.
[22] 刘知秋. 短期电力负荷预测方法研究 [D]. 南京: 东南大学, 2023.
YANG Xueying, QI Qi, LI Qiming, et al. Peak load game scheduling
LIU Zhiqiu. Research on short-term power load forecasting
strategy combining reward mechanism and user willingness[J].
method[D]. Nanjing: Southeast University, 2023.
Transactions of China Electrotechnical Society, 2024, 39(16):
[23] CHEN Z Q, YANG Y, JIANG C D, et al. Enhanced industrial heat
5060–5074.
load forecasting in district networks via a multi-scale fusion ensemble
[14] XUE W, HE H, WANG Y B, et al. SAGN: sparse adaptive gated
deep learning[J]. Expert Systems with Applications, 2025, 272:
graph neural network with graph regularization for identifying dual-
126783.
view brain networks[J]. IEEE Transactions on Neural Networks and
[24] ZHANG K, YANG M W, ZHANG Y M, et al. Error feedback
Learning Systems, 2025, 36(5): 8085–8099.
method (EFM) based dimension synthesis optimisation for four-bar
[15] 谈耀荻, 黄艳国, 刘景锋, 等. 基于 MMoE-CNN-Informer 模型的电
linkage mechanism[J]. Applied Soft Computing, 2023, 144: 110424.
力系统多元负荷长短期时间序列预测 [J]. 电气工程学报, 2025,
[25] 葛璐璐. 基于滑动窗口的一类非负可变权组合预测方法 [D]. 合肥:
20(2): 253–263.
安徽大学, 2019.
TAN Yaodi, HUANG Yanguo, LIU Jingfeng, et al. Long and short-
GE Lulu. A new non-negative variable weight combination
term multivariate load forecasting based on MMoE-CNN-Informer
forecasting method based on sliding window[D]. Hefei: Anhui
model for power systems[J]. Journal of Electrical Engineering, 2025,
University, 2019.
20(2): 253–263.
[16] WANG R F, WU J T, CHENG X, et al. Adaptive expert fusion [26] 于多, 曹燚, 王海荣, 等. 基于 ICEEMDAN-PE 和 IDBO-Informer
model for online wind power prediction[J]. Neural Networks, 2025, 组合模型的短期负荷预测 [J]. 中国电力, 2025, 58(6): 19–32.
184: 107022. YU Duo, CAO Yi, WANG Hairong, et al. Short-term load
[17] 程鹏举, 樊艳芳, 侯俊杰, 等. 基于 Conformer-MoE 的多设备迁移 forecasting based on a combined ICEEMDAN-PE and IDBO-
学习非侵入式负荷分解方法 [J]. 智慧电力, 2026, 54(1): 102–109. Informer model[J]. Electric Power, 2025, 58(6): 19–32.
CHENG Pengju, FAN Yanfang, HOU Junjie, et al. A multi-device [27] 李笑竹, 王维庆. 基于贝叶斯理论的分布鲁棒优化在储能配置上的
transfer learning non-intrusive load disaggregation method based on 应用 [J]. 电网技术, 2022, 46(10): 4001–4011.
conformer-MoE[J]. Smart Power, 2026, 54(1): 102–109. LI Xiaozhu, WANG Weiqing. Application of distributed robust

131