Page 59 - 《中国电力》2026年第5期

随泽远等：基于门控脉冲神经 P 系统模型的概率负荷预测 2026 年第 5 期

11 System Protection and Control, 2024, 52(18): 112–122.
[4] 甘业平, 白云龙, 韩号, 等. 基于动态权重模型组合的短期区域净负
10
荷预测方法 [J]. 电力信息与通信技术, 2025, 23(3): 17–24.
GAN Yeping, BAI Yunlong, HAN Hao, et al. Short-term regional net
9
真实值；
负荷/MW 本文方法； load forecasting method based on dynamic weight model
GSNP；
8
LSTM； combination[J]. Electric Power Information and Communication
Technology, 2025, 23(3): 17–24.
GRU；
7
CBLM； [5] 谭智文, 徐茹枝, 关志涛. 基于差分隐私的个性化联邦电力负荷预
CBGU 测方案 [J]. 电力信息与通信技术, 2024, 22(7): 18–26.
6
0 10 20 30 40 50 TAN Zhiwen, XU Ruzhi, GUAN Zhitao. A personalized federal
采样点/30 min
power load forecasting scheme based on differential privacy[J].
图 8 不同模型的确定性点预测结果（案例 2） Electric Power Information and Communication Technology, 2024,
Fig. 8 Point prediction results of different
22(7): 18–26.
models (case 2)
[6] 庞松岭, 赵雨楠, 唐金锐, 等. 基于充电桩利用率的充电负荷超短期
表 9 确定性点预测模型指标比较（案例 2） 预测方法研究 [J]. 电力科学与技术学报, 2024, 39(1): 115–123.

Table 9 Comparison of indicators for deterministic point
PANG Songling, ZHAO Yunan, TANG Jinrui, et al. A novel ultra
prediction models (case 2)
short-term charging load forecasting method based on usage degree
模型 E MA /kW E MAP /% R 2
of charging piles[J]. Journal of Electric Power Science and
本文方法 71.71 0.884 0.99
Technology, 2024, 39(1): 115–123.
GSNP 79.71 0.986 0.99
[7] 刘舒, 姚尚坤, 周敏, 等. 基于 ICEEMDAN-TA-LSTM 模型的主动
LSTM 83.03 1.037 0.99
配电网短期运行态势预测 [J]. 电力科学与技术学报, 2023, 38(6):
GRU 80.52 1.000 0.99
175–186.
CBLM 171.92 2.088 0.97
LIU Shu, YAO Shangkun, ZHOU Min, et al. Active distribution
CBGU 150.80 1.877 0.97
network operating situation prediction based on ICEEMDAN-TA-
LSTM model[J]. Journal of Electric Power Science and Technology,
精度。此外，未来还可研究极端天气和突发事件
2023, 38(6): 175–186.
下的精确概率预测。
[8] 苏向敬, 朱敏轩, 宇海波, 等. 基于频谱注意力和无交叉联合分位数
回归的海上风电功率超短期概率预测 [J]. 电力系统保护与控制,
参考文献：

2024, 52(21): 103–116.
SU Xiangjing, ZHU Minxuan, YU Haibo, et al. Ultra-short-term
[1] KONG X Y, LI C, ZHENG F, et al. Improved deep belief network
probabilistic forecasting of offshore wind power based on spectral
for short-term load forecasting considering demand-side manage-
attention and non-crossing joint quantile regression[J]. Power System
ment[J]. IEEE Transactions on Power Systems, 2020, 35(2):
Protection and Control, 2024, 52(21): 103–116.
1531–1538.
[9] XU C L, SUN Y J, DUA A, et al. Quantile regression based
[2] 张恒, 郑建勇, 梅飞, 等. 基于 VMD 和辅助任务学习的短期负荷预
probabilistic forecasting of renewable energy generation and building
测方法 [J]. 电力系统保护与控制, 2025, 53(5): 104–112.
electrical load: A state of the art review[J]. Journal of Building
ZHANG Heng, ZHENG Yongjian, MEI Fei, et al. Short-term load Engineering, 2023, 79: 107772.
forecasting method based on VMD and auxiliary task learning[J]. [10] 万灿, 崔文康, 宋永华. 新能源电力系统概率预测: 基本概念与数
Power System Protection and Control, 2025, 53(5): 104–112. 学原理 [J]. 中国电机工程学报, 2021, 41(19): 6493–6508.
[3] 和萍, 刘鑫, 宫智杰, 等. 高比例可再生能源电力系统源荷储联合调 WAN Can, CUI Wenkang, SONG Yonghua. Probabilistic forecasting
峰分层优化运行 [J]. 电力系统保护与控制, 2024, 52(18): 112–122. for power systems with renewable energy sources basic concepts and
HE Ping, LIU Xin, GONG Zhijie, et al. Hierarchical optimization mathematical principles[J]. Proceedings of the CSEE, 2021, 41(19):
operation model for joint peak-load regulation of source-load-storage 6493–6508.
in a high proportion of renewable energy power system[J]. Power [11] 赵洪山, 吴雨晨, 温开云, 等. 基于时空注意力机制的台区多用户短

55