基于模式互相关和时序片段关联机制的多元负荷预测

doi:10.19678/j.issn.1000-3428.0070572

计算机工程 ›› 2026, Vol. 52 ›› Issue (8): 315-326. doi: 10.19678/j.issn.1000-3428.0070572

基于模式互相关和时序片段关联机制的多元负荷预测

李整¹^,², 李志霄¹^,*(), 秦金磊¹^,³, 郭长珍¹

1. 华北电力大学计算机系, 河北保定 071003
2. 华北电力大学河北省能源电力知识计算重点实验室, 河北保定 071003
3. 华北电力大学复杂能源系统智能计算教育部工程研究中心, 河北保定 071003

收稿日期:2024-11-04 修回日期:2025-02-13 出版日期:2026-08-15 发布日期:2025-03-21
通讯作者: 李志霄
作者简介:
李整, 女, 副教授, 主研方向为智能计算、电力系统负荷预测及优化调度
李志霄(通讯作者), 硕士研究生
秦金磊, 副教授
郭长珍, 硕士研究生
基金资助:
河北省自然科学基金(F2014502081); 中央高校基本科研业务费专项资金(2020MS120)

Multiple Loads Forecasting Based on Pattern Cross-Correlation and Temporal Patch Association Mechanism

LI Zheng¹^,², LI Zhixiao¹^,*(), QIN Jinlei¹^,³, GUO Changzhen¹

1. Department of Computer, North China Electric Power University, Baoding 071003, Hebei, China
2. Hebei Key Laboratory of Knowledge Computing for Energy and Power, North China Electric Power University, Baoding 071003, Hebei, China
3. Engineering Research Center of Intelligent Computing for Complex Energy Systems, Ministry of Education, North China Electric Power University, Baoding 071003, Hebei, China

Received:2024-11-04 Revised:2025-02-13 Online:2026-08-15 Published:2025-03-21
Contact: LI Zhixiao

摘要/Abstract

摘要：

针对综合能源系统(IES)负荷噪声大、波动性强、周期信息提取困难问题, 提出一种基于模式互相关和时序片段关联(PCC-TPA)机制的多元负荷预测方法。通过互相关函数分析外界影响因素与多元负荷间的交叉滞后关系, 确定最相关的时滞并据此重构和嵌入数据。在此基础上, 提出PCC机制, 根据数据的变化趋势将其抽象为模式数据, 以减轻数据波动与噪声的影响, 基于互相关理论识别和提取关键时刻及周期信息; 同时, 设计时序片段关联机制, 将序列划分为多个子序列, 基于互信息方法分析和筛选子序列, 增强模型对序列局部连续性信息的捕获能力。在美国亚利桑那州立大学坦佩校区综合能源系统数据集上进行多项消融实验和对比实验。实验结果表明, 交叉滞后分析重构的数据有效提升了模型的预测精度, PCC-TPA机制分别增强模型的关键时刻识别能力和局部信息捕捉能力。所提方法在多项评估指标上优于5种主流预测模型, 具有较高的预测精度。

关键词: 多元负荷预测, 交叉滞后分析, 序列分解, 序列模式化, 互相关分析, 深度学习

Abstract:

To address the challenges of high noise levels, strong volatility, and difficulties in extracting periodic information from the loads of Integrated Energy Systems (IES), a multivariate load forecasting method based on Pattern Cross-Correlation and Temporal Patch Association (PCC-TPA) mechanisms is proposed. This method analyzes the cross-lag relationships between external influencing factors and multivariate loads using a cross-correlation function to determine the most relevant time lags for data reconstruction and embedding. Building on this, a PCC mechanism is introduced to abstract data into patterns based on their variation trends, thus mitigating the effects of fluctuations and noise. This is followed by the identification and extraction of key moments and periodic information based on the cross-correlation theory. A TPA mechanism is designed to divide a sequence into multiple subsequences using mutual information methods to analyze and filter the subsequences, thereby enhancing the ability of the model to capture local continuity information in sequences. Multiple ablation and comparison experiments are conducted using the comprehensive energy system dataset from the Arizona state university, Tempe campus. The ablation experimental results show that the data reconstructed through cross-lag analysis effectively improved the prediction accuracy of the model. The PCC-TPA mechanisms enhance the ability of the model to identify key moments and capture local information, respectively. The comparison of the experimental results indicates that the proposed method outperforms five mainstream prediction models in terms of multiple evaluation metrics, demonstrating a higher prediction accuracy.

Key words: multiple loads forecasting, cross-lagged analysis, series decomposition, series pattern, cross-correlation analysis, deep learning

李整, 李志霄, 秦金磊, 郭长珍. 基于模式互相关和时序片段关联机制的多元负荷预测[J]. 计算机工程, 2026, 52(8): 315-326.

LI Zheng, LI Zhixiao, QIN Jinlei, GUO Changzhen. Multiple Loads Forecasting Based on Pattern Cross-Correlation and Temporal Patch Association Mechanism[J]. Computer Engineering, 2026, 52(8): 315-326.

https://www.ecice06.com/CN/Y2026/V52/I8/315

图/表 20

图1 多元负荷与影响因素相关性

Fig.1 Correlation between multiple loads and influencing factors

图2 电负荷交叉滞后分析

Fig.2 Cross-lagged analysis of electrical loads

图3 数据处理与嵌入

Fig.3 Data processing and embedding

图4 模型总体结构

Fig.4 Overall structure of the model

图5 模式互相关机制结构

Fig.5 Structure of the pattern cross-correlation mechanism

图6 波动数据验证

Fig.6 Validation of fluctuating data

图7 数值分布不均数据验证

Fig.7 Validation of non-uniformly distributed numerical data

图8 时序片段关联

Fig.8 Temporal patch association

图9 自注意力机制权重分布

Fig.9 Weight distribution of the self-attention mechanism

图10 PCC机制权重分布

Fig.10 Weight distribution of the PCC mechanism

图11 电负荷夏季预测结果

Fig.11 Summer prediction results of electrical loads

图12 电负荷冬季预测结果

Fig.12 Winter prediction results of electrical loads

图13 冷负荷预测结果

Fig.13 Prediction results of cooling load

图14 热负荷预测结果

Fig.14 Prediction results of heating load

参考文献 29

1	史佳琪, 谭涛, 郭经, 等. 基于深度结构多任务学习的园区型综合能源系统多元负荷预测. 电网技术, 2018, 42 (3): 698- 707.
	SHI J Q , TAN T , GUO J , et al. Multi-task learning based on deep architecture for various types of load forecasting in regional energy system integration. Power System Technology, 2018, 42 (3): 698- 707.
2	叶剑华, 曹旌, 杨理, 等. 基于变分模态分解和多模型融合的用户级综合能源系统超短期负荷预测. 电网技术, 2022, 46 (7): 2610- 2622.
	YE J H , CAO J , YANG L , et al. Ultra short-term load forecasting of user level integrated energy system based on variational mode decomposition and multi-model fusion. Power System Technology, 2022, 46 (7): 2610- 2622.
3	LIU Y J , LI Y , LIU G L , et al. Review of multiple load forecasting method for integrated energy system. Frontiers in Energy Research, 2023, 11, 101- 120.
4	ZHANG J Z , DONG H J , ZHANG W Y , et al. Review and prospect of data-driven techniques for load forecasting in integrated energy systems. Applied Energy, 2022, 321, 119269. doi: 10.1016/j.apenergy.2022.119269
5	邓带雨, 李坚, 张真源, 等. 基于EEMD-GRU-MLR的短期电力负荷预测. 电网技术, 2020, 44 (2): 593- 602.
	DENG D Y , LI J , ZHANG Z Y , et al. Short-term electric load forecasting based on EEMD-GRU-MLR. Power System Technology, 2020, 44 (2): 593- 602.
6	ANTONIO B , PIERLUIGI C , PASQUALE F D , et al. Multivariate quantile regression for short-term probabilistic load forecasting. IEEE Transactions on Power Systems, 2020, 35 (1): 628- 638. doi: 10.1109/TPWRS.2019.2924224
7	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. 中国电机工程学报, 2021, 41 (19): 6493- 6509.
	WAN C , CUI W K , SONG Y H . Probabilistic forecasting for power systems with renewable energy sources: basic concepts and mathematical principles. Proceedings of CSEE, 2021, 41 (19): 6493- 6509.
8	JASON R , ETIENNE S . A comparison of prediction and forecasting artificial intelligence models to estimate the future energy demand in a district heating system. Energy, 2023, 269, 126661. doi: 10.1016/j.energy.2023.126661
9	史佳琪, 张建华. 基于多模型融合Stacking集成学习方式的负荷预测方法. 中国电机工程学报, 2019, 39 (14): 4032- 4041.
	SHI J Q , ZHANG J H . Load forecasting based on multi-model by Stacking ensemble learning. Proceedings of the CSEE, 2019, 39 (14): 4032- 4041.
10	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. 中国电机工程学报, 2023, 43 (23): 9072- 9084.
	WANG W , FENG B , HUANG G , et al. Short-term net load forecasting based on self-attention encoder and deep neural network. Proceedings of the CSEE, 2023, 43 (23): 9072- 9084.
11	李丹, 孙光帆, 缪书唯, 等. 基于多维时序信息融合的短期电力负荷预测方法. 中国电机工程学报, 2023, 43 (S1): 94- 106.
	LI D , SUN G F , MIAO S W , et al. A short-term power load forecasting method based on multidimensional temporal information fusion. Proceedings of the CSEE, 2023, 43 (S1): 94- 106.
12	NADA M , HAMID O , ISMAEL J . Short-term electric load forecasting using an EMD-BI-LSTM approach for smart grid energy management system. Energy and Buildings, 2023, 288, 113022. doi: 10.1016/j.enbuild.2023.113022
13	LIU H X , SHEN X D , TANG X S , et al. Day-ahead electricity price probabilistic forecasting based on SHAP feature selection and LSTNet quantile regression. Energies, 2023, 16 (13): 101- 109.
14	王琛, 王颖, 郑涛, 等. 基于ResNet-LSTM网络和注意力机制的综合能源系统多元负荷预测. 电工技术学报, 2022, 37 (7): 1789- 1799.
	WANG C , WANG Y , ZHENG T , et al. Multi-energy load forecasting in integrated energy system based on ResNet-LSTM network and attention mechanism. Transactions of China Electrotechnical Society, 2022, 37 (7): 1789- 1799.
15	CAI C C , LI Y J , SU Z H , et al. Short-term electrical load forecasting based on VMD and GRU-TCN hybrid network. Applied Sciences, 2022, 12 (13): 6647. doi: 10.3390/app12136647
16	CHEN W H , HAN G J , ZHU H B , et al. Short-term load forecasting with an ensemble model using densely residual block and Bi-LSTM based on the attention mechanism. Sustainability, 2022, 14 (24): 16433. doi: 10.3390/su142416433
17	QIN Y , LU Z Y , LIU H , et al. An improved feature-time Transformer encoder-Bi-LSTM for short-term forecasting of user-level integrated energy loads. Energy[WT《Times New Roman》] & Buildings, 2023, 297, 113396.
18	WU H X, XU J H, WANG J M, et al. Autoformer: decomposition Transformers with auto-correlation for long-term series forecasting[EB/OL]. [2024-10-01]. https://arxiv.org/pdf/2106.13008.
19	CEN S F , LIM C G . Multi-task learning of the PatchTCN-TST model for short-term multi-load energy forecasting considering indoor environments in a smart building. IEEE Access, 2024, 12, 19553- 19568. doi: 10.1109/ACCESS.2024.3355448
20	鲁斌, 霍泽健, 俞敏. 基于LSTNet-Skip的综合能源系统多元负荷超短期预测. 中国电机工程学报, 2023, 43 (6): 2273- 2283.
	LU B , HUO Z J , YU M . Multi load ultra short-term forecasting of integrated energy system based on LSTNet-Skip. Proceedings of the CSEE, 2023, 43 (6): 2273- 2283.
21	ZUO C H , WANG J L , LIU M P , et al. An ensemble framework for short-term load forecasting based on TimesNet and TCN. Energies, 2023, 16 (14): 5330. doi: 10.3390/en16145330
22	石卓见, 冉启武, 徐福聪. 基于聚合二次模态分解及Informer的短期负荷预测. 电网技术, 2023, 48 (6): 2574- 2583.
	SHI Z J , RAN Q W , XU F C . Short-term load forecasting based on aggregated secondary decomposition and Informer. Power System Technology, 2023, 48 (6): 2574- 2583.
23	NIU D X , YU M , SUN L J , et al. Short-term multi-energy load forecasting for integrated energy systems based on CNN-BiGRU optimized by attention mechanism. Applied Energy, 2022, 313, 118801. doi: 10.1016/j.apenergy.2022.118801
24	HE F F , ZHOU J Z , FENG Z K , et al. A hybrid short-term load forecasting model based on variational mode decomposition and long short-term memory networks considering relevant factors with Bayesian optimization algorithm. Applied Energy, 2019, 237, 103- 116. doi: 10.1016/j.apenergy.2019.01.055
25	李云松, 张智晟. 基于GRU-TGTransformer的综合能源系统多元负荷短期预测. 电力系统保护与控制, 2023, 51 (15): 33- 41.
	LI Y S , ZHANG Z S . Multi load short-term forecasting of an integrated energy system based on a GRU-TGTransformer. Power System Protection and Control, 2023, 51 (15): 33- 41.
26	李鹏, 罗湘淳, 孟庆伟, 等. 基于Spearman相关性阈值寻优和VMD-LSTM的用户级综合能源系统超短期负荷预测. 全球能源互联网, 2024, 7 (4): 406- 420.
	LI P , LUO X C , MENG Q W , et al. Ultra short-term load forecasting of user level integrated energy system based on Spearman threshold optimization and variational mode decomposition, and long short-term memory. Journal of Global Energy Interconnection, 2024, 7 (4): 406- 420.
27	张大海, 杨宇辰, 刘艳梅, 等. 基于EMD与Spearman相关系数的混合直流线路纵联保护方法. 电力系统保护与控制, 2021, 49 (9): 1- 11.
	ZHANG D H , YANG Y C , LIU Y M , et al. Hybrid HVDC line pilot protection method based on EMD and Spearman correlation coefficient. Power System Protection and Control, 2021, 49 (9): 1- 11.
28	ZHENG X Y , CHEN Q Q , SUN M C , et al. Exploring the influence of environmental indicators and forecasting influenza incidence using ARIMAX models. Frontiers in Energy Research, 2024, 12, 1441240.
29	管奔, 臧勇, 吴迪平. 基于互相关函数的轧机振动传递时间滞后分析[C]//第七届中国钢铁年会. 北京: 冶金工业出版社, 2009: 345-350.
	GUAN B, ZANG Y, WU D P. Study of delay time of transfer of rolling mill vibration based on cross-correlation function[C]//Proceedings of the 7th China Steel Annual Conference. Beijing: Metallurgical Industry Press, 2009: 345-350. (in Chinese)

[1]	王忠美, 聂芃轩, 刘建华, 吴海波, 郑良. 基于多约束模态不变图卷积融合网络的轴承故障诊断[J]. 计算机工程, 2026, 52(8): 422-430.
[2]	樊帅, 陈永辉, 谯旭, 肖立伟. 基于几何适应与多尺度门控融合的二维障碍物流场预测[J]. 计算机工程, 2026, 52(8): 376-389.
[3]	胡财富, 魏波, 任芮彬. 空间优化视角下基于增量学习的网络流量分类[J]. 计算机工程, 2026, 52(8): 138-148.
[4]	赵莹莹, 朱率率. 知识图谱构建技术研究综述[J]. 计算机工程, 2026, 52(8): 84-100.
[5]	冯嘉琦, 高见. 面向网络威胁情报领域的信息抽取综述[J]. 计算机工程, 2026, 52(7): 56-75.
[6]	魏方达, 刘淼, 孙毅, 王晶, 赵胜辉. 音视频深度伪造与鉴伪综述[J]. 计算机工程, 2026, 52(7): 40-55.
[7]	梁子仪, 王子豪, 刘天权, 李丽萍, 朱远飞, 鹿存跃. PASRGAN: 基于并联注意力机制的超分辨率病理图像生成算法[J]. 计算机工程, 2026, 52(7): 189-198.
[8]	胡康源, 郭涛, 穆楠. 基于自注意力机制和动态掩膜机制的文物图像修复方法[J]. 计算机工程, 2026, 52(6): 179-188.
[9]	孙海峰, 姚俊萍, 李晓军, 刘延飞, 辜弘炀. 短期动作预测深度学习方法综述[J]. 计算机工程, 2026, 52(6): 31-52.
[10]	李学相, 郑永利, 张怡泽, 段鹏松. 基于机器学习与预训练模型的流量分析方法综述[J]. 计算机工程, 2026, 52(6): 53-67.
[11]	肖泽秋, 李勇, 王霞. 基于PBI-CLA模型的糖尿病患者血糖浓度预测[J]. 计算机工程, 2026, 52(6): 382-390.
[12]	李亦然, 聂宏宾, 杨紫骞, 卞春江. 密集星场下空间暗弱群组目标检测方法[J]. 计算机工程, 2026, 52(6): 149-159.
[13]	许旻辰, 屈丹, 司念文, 彭思思, 陈雅淇. 社交媒体虚假信息检测技术研究综述[J]. 计算机工程, 2026, 52(5): 60-80.
[14]	李辉, 刘佳煜, 徐雅萍. 面向医学图像分割的深度学习模型架构与性能评估方法综述[J]. 计算机工程, 2026, 52(5): 81-94.
[15]	成彬, 赵彬兵, 雷华, 何博. 基于双目视觉的钢筋绑扎节点定位方法[J]. 计算机工程, 2026, 52(4): 433-445.

选择文件类型/文献管理软件名称

选择包含的内容

基于模式互相关和时序片段关联机制的多元负荷预测

Multiple Loads Forecasting Based on Pattern Cross-Correlation and Temporal Patch Association Mechanism

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 20

参考文献 29

相关文章 15

编辑推荐

Metrics

本文评价

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于模式互相关和时序片段关联机制的多元负荷预测

Multiple Loads Forecasting Based on Pattern Cross-Correlation and Temporal Patch Association Mechanism

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 20

参考文献 29

相关文章 15

编辑推荐

Metrics

本文评价