基于门控脉冲神经P系统模型的概率负荷预测

doi:10.11930/j.issn.1004-9649.202505075

中国电力 ›› 2026, Vol. 59 ›› Issue (5): 46-56.DOI: 10.11930/j.issn.1004-9649.202505075

• 有源配电网安全高效运行与协同调控关键技术 • 上一篇下一篇

基于门控脉冲神经P系统模型的概率负荷预测

随泽远¹(), 王军¹(), 彭宏¹(), 王德林², 宋戈³

1. 西华大学电气与电子信息学院，四川成都　610039
2. 西南交通大学电气工程学院，四川成都　610031
3. 国网四川省电力公司成都供电公司，四川成都　610021

收稿日期:2025-05-27 修回日期:2026-04-08 发布日期:2026-05-15 出版日期:2026-05-28
作者简介:
随泽远（1998），男，硕士研究生，从事电力系统负荷预测研究，E-mail：suizeyuan@stu.xhu.edu.cn
王军（1966），女，通信作者，博士，教授，从事分布式电源和微电网管控技术、电力电子节能技术、新型交流电机控制技术、智能控制技术、膜计算等研究，E-mail：wj.xhu@outlook.com
彭宏（1966），男，博士，教授，从事模式识别与图像处理、生物启发的新型计算模型等研究，E-mail：ph.xhu@hotmail.com
基金资助:
国家自然科学基金资助项目（62176216）。

Probabilistic load prediction based on gated spiking neural P system model

SUI Zeyuan¹(), WANG Jun¹(), PENG Hong¹(), WANG Delin², SONG Ge³

1. School of Electrical Engineering and Electronic Information, Xihua University, Chengdu 610039, China
2. School of Electrical Engineering, Southwest Jiaotong University, Chengdu 610031, China
3. State Grid Sichuan Electric Power Company Chengdu Power Supply Bureau, Chengdu 610021, China

Received:2025-05-27 Revised:2026-04-08 Online:2026-05-15 Published:2026-05-28
Supported by:
This work is supported by the National Natural Science Foundation of China (No.62176216).

摘要/Abstract

摘要：

传统的确定性负荷预测无法提供负载的不确定性信息，概率负荷预测能够生成预测值不确定性的概率分布，为电网调度决策提供更丰富的信息。为了进一步提高概率负荷预测的精度，提出了一种包含最小绝对收缩和选择算子（least absolute shrinkage and selection operator，LASSO）及门控脉冲神经P系统（gated spiking neural P system，GSNP）的LASSO-GSNP模型。首先，运用LASSO从最低温度、最高温度、平均温度、平均湿度和降雨量等外部特征中提取关键特征；随后，提出了改进的GSNP模型实现概率负荷预测，以提升长时间序列预测的性能。使用2个不同尺度的长时间序列数据集作为算例，结果表明，所提模型在预测精度指标和预测区间质量上均优于其他几种典型模型。

关键词: 概率负荷预测, 最小绝对收缩和选择算子, 深度神经网络, 分位数回归, 门控脉冲神经P系统

Abstract:

Conventional deterministic load forecasting fails to provide uncertainty information of loads, while probabilistic load forecasting can generate probability distributions of predicted value uncertainty, thus providing comprehensive information for power grid dispatch decisions. In order to further improve the accuracy of probabilistic load forecasting, this paper proposes a model, which incorporates the least absolute shrinkage and selection operator (LASSO) and gated spiking neural P system (GSNP). Firstly, LASSO is used to extract the key features from external features such as minimum temperature, maximum temperature, average temperature, average humidity and precipitation. Subsequently, an improved GSNP model is developed to implement probabilistic load forecasting, enhancing the performance of long-term time-series forecasting. The case study using two long-term time-series datasets at different scales shows that the proposed model outperforms several other typical models in terms of both prediction accuracy and prediction interval quality.

Key words: probabilistic load forecasting, least absolute shrinkage and selection operator, deep neural network, quantile regression, gated spiking neural P system

随泽远, 王军, 彭宏, 王德林, 宋戈. 基于门控脉冲神经P系统模型的概率负荷预测[J]. 中国电力, 2026, 59(5): 46-56.

SUI Zeyuan, WANG Jun, PENG Hong, WANG Delin, SONG Ge. Probabilistic load prediction based on gated spiking neural P system model[J]. Electric Power, 2026, 59(5): 46-56.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.202505075

https://www.electricpower.com.cn/CN/Y2026/V59/I5/46

图/表 17

参考文献 30

1	KONG X Y, LI C, ZHENG F, et al. Improved deep belief network for short-term load forecasting considering demand-side management[J]. IEEE Transactions on Power Systems, 2020, 35 (2): 1531- 1538.
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测方法[J]. 电力系统保护与控制, 2025, 53 (5): 104- 112.
	ZHANG Heng, ZHENG Yongjian, MEI Fei, et al. Short-term load forecasting method based on VMD and auxiliary task learning[J]. Power System Protection and Control, 2025, 53 (5): 104- 112.
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峰分层优化运行[J]. 电力系统保护与控制, 2024, 52 (18): 112- 122.
	HE Ping, LIU Xin, GONG Zhijie, et al. Hierarchical optimization operation model for joint peak-load regulation of source-load-storage in a high proportion of renewable energy power system[J]. Power 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 load forecasting method based on dynamic weight model combination[J]. Electric Power Information and Communication Technology, 2025, 23 (3): 17- 24.
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测方案[J]. 电力信息与通信技术, 2024, 22 (7): 18- 26.
	TAN Zhiwen, XU Ruzhi, GUAN Zhitao. A personalized federal power load forecasting scheme based on differential privacy[J]. Electric Power Information and Communication Technology, 2024, 22 (7): 18- 26.
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预测方法研究[J]. 电力科学与技术学报, 2024, 39 (1): 115- 123.
	PANG Songling, ZHAO Yunan, TANG Jinrui, et al. A novel ultra short-term charging load forecasting method based on usage degree of charging piles[J]. Journal of Electric Power Science and Technology, 2024, 39 (1): 115- 123.
7	刘舒, 姚尚坤, 周敏, 等. 基于ICEEMDAN-TA-LSTM模型的主动配电网短期运行态势预测[J]. 电力科学与技术学报, 2023, 38 (6): 175- 186.
	LIU Shu, YAO Shangkun, ZHOU Min, et al. Active distribution network operating situation prediction based on ICEEMDAN-TA-LSTM model[J]. Journal of Electric Power Science and Technology, 2023, 38 (6): 175- 186.
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, 2024, 52 (21): 103- 116.
	SU Xiangjing, ZHU Minxuan, YU Haibo, et al. Ultra-short-term probabilistic forecasting of offshore wind power based on spectral attention and non-crossing joint quantile regression[J]. Power System Protection and Control, 2024, 52 (21): 103- 116.
9	XU C L, SUN Y J, DUA A, et al. Quantile regression based probabilistic forecasting of renewable energy generation and building electrical load: A state of the art review[J]. Journal of Building Engineering, 2023, 79, 107772.
10	万灿, 崔文康, 宋永华. 新能源电力系统概率预测: 基本概念与数学原理[J]. 中国电机工程学报, 2021, 41 (19): 6493- 6508.
	WAN Can, CUI Wenkang, SONG Yonghua. Probabilistic forecasting for power systems with renewable energy sources basic concepts and mathematical principles[J]. Proceedings of the CSEE, 2021, 41 (19): 6493- 6508.
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期负荷预测[J]. 电工技术学报, 2024, 39 (7): 2104- 2115.
	ZHAO Hongshan, WU Yuchen, WEN Kaiyun, et al. Short-term load forecasting for multiple customers in a station area based on spatial-temporal attention mechanism[J]. Transactions of China Electrotechnical Society, 2024, 39 (7): 2104- 2115.
12	ZHANG D, WANG S, LIANG Y. A novel combined model for probabilistic load forecasting based on deep learning and improved optimizer[J]. Energy, 2023, 264, 126172.
13	赵佩, 代业明. 基于实时电价和加权灰色关联投影的SVM电力负荷预测[J]. 电网技术, 2020, 44 (4): 1325- 1332.
	ZHAO Pei, DAI Yeiming. Power load forecasting of SVM based on real-time price and weighted grey relational projection algorithm[J]. Power System Technology, 2020, 44 (4): 1325- 1332.
14	RUNGE J, SALOUX E. A comparison of prediction and forecasting artificial intelligence models to estimate the future energy demand in a district heating system[J]. Energy, 2023, 269, 126661.
15	DANG S, PENG L, ZHAO J, et al. A quantile regression random forest-based short-term load probabilistic forecasting method[J]. Energies, 2022, 15 (2): 663.
16	ZHANG W, HE Y, YANG S. Day-ahead load probability density forecasting using monotone composite quantile regression neural network and kernel density estimation[J]. Electric Power Systems Research, 2021, 201, 107551.
17	邵必林, 纪丹阳. 基于VMD-SE的电力负荷分量的多特征短期预测[J]. 中国电力, 2024, 57 (4): 162- 170.
	SHAO Bilin, JI Danyang. Multi-feature short-term prediction of power load components based on VMD-SE[J]. Applied Energy, 2024, 57 (4): 162- 170.
18	杨胡萍, 余阳, 汪超, 等. 基于VMD-CNN-BIGRU的电力系统短期负荷预测[J]. 中国电力, 2022, 55 (10): 71- 76.
	YANG Huping, YU Yang, WANG Chao, et al. Short-term load forecasting of power system based on VMD-CNN-BIGRU[J]. Electric Power, 2022, 55 (10): 71- 76.
19	LUO L, DONG J, KONG W, et al. Short-term probabilistic load forecasting using quantile regression neural network with accumulated hidden layer connection structure[J]. IEEE Transactions on Industrial Informatics, 2024, 20 (4): 5818- 5828.
20	ZHANG W, QUAN H, SRINIVASAN D. An improved quantile regression neural network for probabilistic load forecasting[J]. IEEE Transactions on Smart Grid, 2019, 10 (4): 4425- 4434.
21	王炜, 冯斌, 黄刚, 等. 基于自注意力编码器和深度神经网络的短期净负荷预测[J]. 中国电机工程学报, 2023, 43 (23): 9072- 9084.
	WANG Wei, FENG Bin, HUANG Gang, et al. Short-term net load forecasting based on self-attention encoder and deep neural network[J]. Proceedings of the CSEE, 2023, 43 (23): 9072- 9084.
22	李科, 潘庭龙, 许德智. 基于MSCNN-BiGRU-Attention的短期电力负荷预测[J]. 中国电力, 2025, 58 (6): 10- 18.
	LI Ke, PAN Tinglong, XU Dezhi. Short-term power load forecasting based on MSCNN-BiGRU-Attention[J]. Electric Power, 2025, 58 (6): 10- 18.
23	ANEIROS G, NOVO S, VIEU P. Variable selection in functional regression models: A review[J]. Journal of Multivariate Analysis, 2022, 188, 104871.
24	TIBSHIRANI R. Regression shrinkage and selection via the lasso[J]. Journal of the Royal Statistical Society, 1996, 58 (1): 267- 288.
25	LU S, XU Q, JIANG C, et al. Probabilistic load forecasting with a non-crossing sparse-group lasso-quantile regression deep neural network[J]. Energy, 2022, 242, 122955.
26	LIU S M, CHEN H W, LIU P X, et al. A novel electricity load forecasting based on probabilistic least absolute shrinkage and selection operator-quantile regression neural network[J]. International Journal of Hydrogen Energy, 2023, 48, 34486- 34500.
27	LIU Q, LONG L, PENG H, et al. Gated spiking neural P systems for time series forecasting[J]. IEEE Transactions on Neural Networks and Learning Systems, 2023, 34 (9): 6227- 6236.
28	IONESCU M, PAUN G, YOKOMORI T, et al. Spiking neural P systems[J]. Fundamenta Informaticae, 2006, 71 (2-3): 279- 308.
29	EMMERT-STREIB F, DEHMER M. High-dimensional LASSO-based computational regression models: regularization, shrinkage, and selection[J]. Machine Learning and Knowledge Extraction, 2019, 1 (1): 359- 383.
30	陈轩伟. 基于BP-QR模型的负荷区间预测[J]. 电气技术, 2022, 23 (4): 14- 17.
	CHEN Xuanwei. Load interval forecasting based on BP-QR model[J]. Electrical Engineering, 2022, 23 (4): 14- 17.

模型	参数
模型共同参数	滞后值：1，神经元个数：24，学习率：0.001，训练轮数：25，优化器：Adam，激活函数：RuLU
CNN	卷积层：1，卷积核数量：32，卷积核大小：1
LASSO	正则化权重：0.001

模型	参数
模型共同参数	滞后值：1，神经元个数：24，学习率：0.001，训练轮数：25，优化器：Adam，激活函数：RuLU
CNN	卷积层：1，卷积核数量：32，卷积核大小：1
LASSO	正则化权重：0.001

置信度/%	本文方法			GSNP			LSTM			GRU			CBLM			CBGU
置信度/%	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS
90	91.57	0.0835	–519.18	92.93	0.1110	–690.49	89.57	0.1126	–700.37	96.14	0.2104	–1307.95	96.00	0.1570	–976.17	94.45	0.1283	–797.67
85	86.79	0.0744	–473.50	88.71	0.0997	–631.03	87.71	0.1007	–670.23	93.93	0.1621	–1127.10	90.75	0.1381	–859.26	90.80	0.1179	–733.42
80	81.71	0.0602	–332.79	85.00	0.0806	–445.15	78.50	0.0797	–440.48	83.07	0.1265	–698.76	81.90	0.1149	–714.61	83.50	0.1033	–642.51

置信度/%	本文方法			GSNP			LSTM			GRU			CBLM			CBGU
置信度/%	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS
90	91.57	0.0835	–519.18	92.93	0.1110	–690.49	89.57	0.1126	–700.37	96.14	0.2104	–1307.95	96.00	0.1570	–976.17	94.45	0.1283	–797.67
85	86.79	0.0744	–473.50	88.71	0.0997	–631.03	87.71	0.1007	–670.23	93.93	0.1621	–1127.10	90.75	0.1381	–859.26	90.80	0.1179	–733.42
80	81.71	0.0602	–332.79	85.00	0.0806	–445.15	78.50	0.0797	–440.48	83.07	0.1265	–698.76	81.90	0.1149	–714.61	83.50	0.1033	–642.51

模型	平均 PL值	0.25分位数PL值	0.5分位数PL值	0.75分位数PL值
本文方法	47.83	53.43	65.74	53.13
GSNP	53.43	56.09	72.61	59.42
LSTM	58.40	67.58	77.21	63.72
GRU	75.50	86.69	103.93	83.08
CBLM	85.62	100.63	120.42	89.45
CBGU	76.23	85.16	95.82	78.93

基于门控脉冲神经P系统模型的概率负荷预测

Probabilistic load prediction based on gated spiking neural P system model

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 17

参考文献 30

相关文章 8

编辑推荐

Metrics

特征	相关度/%
最低温度	69.20
最高温度	17.72
平均温度	12.21
平均湿度	0.14
降雨量	0.71

模型	E_MA/kW	E_MAP/%	R²
本文方法	127.60	2.003	0.99
GSNP	144.03	2.221	0.98
LSTM	152.97	2.442	0.98
GRU	198.17	3.165	0.97
CBLM	257.40	4.014	0.96
CBGU	210.98	3.262	0.97

特征	相关度/%
干球湿度	0.470
露点温度	67.550
湿球温度	0.850
湿度	0.073
电价	31.040

[1]	陶军, 钟鸣, 张艺, 刘锋, 武玉珠, 夏振兴. 基于网络流算法和深度神经网络的智能变电站二次系统故障定位方法[J]. 中国电力, 2025, 58(9): 68-78.
[2]	桂前进, 徐文法, 李晓阳, 罗利荣, 叶海峰, 王正风. 基于时序分解与共形分位数回归的超短期光伏功率区间预测[J]. 中国电力, 2025, 58(5): 21-32.
[3]	陈中飞, 赵越, 蔡秋娜, 张乔榆, 王泽林, 戴晓娟, 陈雨果. 基于净负荷预测误差统计的电力系统爬坡能力充裕度评估[J]. 中国电力, 2024, 57(5): 50-60.
[4]	阎洁, 杨佳琳, 王航宇, 卢姣阳, 刘永前, 张磊. 基于风况预测误差自适应的海上风电场尾流偏转控制方法[J]. 中国电力, 2024, 57(3): 190-196.
[5]	陈来恩, 曾小勇, 曾子豪, 成采辰, 孙耀科. 基于物理信息与深度神经网络的锂离子电池温度预测[J]. 中国电力, 2024, 57(11): 18-25.
[6]	唐旭辰, 潮铸, 段秦尉, 苏炳洪, 陈卉灿. 基于分层测量数据的高压变电站概率负荷预测方法[J]. 中国电力, 2023, 56(8): 143-150.
[7]	苏向敬, 宇海波, 符杨, 田书欣, 李海瑜, 耿福海. 基于DALSTM和联合分位数损失的海上风电功率概率预测[J]. 中国电力, 2023, 56(11): 10-19.
[8]	沙骏, 徐雨森, 刘冲冲, 冯定东, 胥峥, 臧海祥. 基于变分模态分解和分位数卷积-循环神经网络的短期风功率预测[J]. 中国电力, 2022, 55(12): 61-68.

模型	平均 PL值	0.25分位数PL值	0.5分位数PL值	0.75分位数PL值
本文方法	26.40	29.21	36.37	29.31
GSNP	29.11	31.95	39.93	32.78
LSTM	29.85	35.10	41.12	30.29
GRU	29.37	34.23	40.06	31.47
CBLM	56.50	56.34	87.09	66.72
CBGU	54.56	55.38	74.21	65.02

模型	平均 PL值	0.25分位数PL值	0.5分位数PL值	0.75分位数PL值
本文方法	26.40	29.21	36.37	29.31
GSNP	29.11	31.95	39.93	32.78
LSTM	29.85	35.10	41.12	30.29
GRU	29.37	34.23	40.06	31.47
CBLM	56.50	56.34	87.09	66.72
CBGU	54.56	55.38	74.21	65.02

置信度/%	本文方法			GSNP			LSTM			GRU			CBLM			CBGU
置信度/%	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS
90	93.64	0.0407	–325.59	96.07	0.0502	–402.18	96.57	0.0512	–410.13	97.21	0.0569	–455.07	92.10	0.0604	–483.84	95.50	0.0938	–751.55
85	91.43	0.0374	–301.38	93.93	0.0440	–371.82	92.14	0.0418	–380.77	96.00	0.0504	–398.28	91.40	0.0514	–411.77	92.95	0.0784	–627.92
80	83.29	0.0307	–218.72	84.00	0.0343	–244.17	84.93	0.0368	–261.82	92.36	0.0430	–305.97	90.10	0.0473	–378.87	87.85	0.0724	–580.22

置信度/%	本文方法			GSNP			LSTM			GRU			CBLM			CBGU
置信度/%	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS	X_PICP	X_PINAW	X_AIS
90	93.64	0.0407	–325.59	96.07	0.0502	–402.18	96.57	0.0512	–410.13	97.21	0.0569	–455.07	92.10	0.0604	–483.84	95.50	0.0938	–751.55
85	91.43	0.0374	–301.38	93.93	0.0440	–371.82	92.14	0.0418	–380.77	96.00	0.0504	–398.28	91.40	0.0514	–411.77	92.95	0.0784	–627.92
80	83.29	0.0307	–218.72	84.00	0.0343	–244.17	84.93	0.0368	–261.82	92.36	0.0430	–305.97	90.10	0.0473	–378.87	87.85	0.0724	–580.22

模态框（Modal）标题

基于门控脉冲神经P系统模型的概率负荷预测

Probabilistic load prediction based on gated spiking neural P system model

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 17

参考文献 30

相关文章 8

编辑推荐

Metrics