Short-term Load Forecasting Based on Temporal Convolutional Network and Transfer Learning

LUO Tongtong; TANG Zhiyuan; TANG Yikun; LIU Junyong

doi:10.19725/j.cnki.1007-2322.2024.0072

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

Short-term Load Forecasting Based on Temporal Convolutional Network and Transfer Learning

更新时间：2026-06-02

- Short-term Load Forecasting Based on Temporal Convolutional Network and Transfer Learning
- Vol. 43, Issue 3, Pages: 422-432(2026)
- 作者机构：
  
  四川大学电气工程学院,四川省,成都市,610065
- 作者简介：
- 基金信息：
- DOI：10.19725/j.cnki.1007-2322.2024.0072
  CLC： 3
- Published：2026
- 稿件说明：
移动端阅览
LUO Tongtong, TANG Zhiyuan, TANG Yikun, et al. Short-term Load Forecasting Based on Temporal Convolutional Network and Transfer Learning[J]. 2026, 43(3): 422-432.
DOI：

LUO Tongtong, TANG Zhiyuan, TANG Yikun, et al. Short-term Load Forecasting Based on Temporal Convolutional Network and Transfer Learning[J]. 2026, 43(3): 422-432. DOI： 10.19725/j.cnki.1007-2322.2024.0072.

摘要

短期负荷预测对于优化能源供需平衡、提高电力系统运行效率至关重要。部分地区由于缺乏相关负荷历史数据集，难以使用传统的机器学习预测方法。针对数据稀缺情况下的短期负荷预测问题，提出一种基于时间卷积网络（temporal convolutional network

TCN）与迁移学习的混合预测模型。首先，为了准确度量源域的可迁移性，提出一种综合考虑数据间距离与特征分布差异的源域选择方法。该方法通过结合马氏距离与最大均值差异，筛选出与目标域具有高度相似性的源域。其次，利用雪消融优化器（snow ablation optimizer

SAO）对TCN模型的超参数进行优化，建立SAO-TCN迁移模型，以提高整体的预测效果。最后，使用实际数据集对所提混合模型进行性能评估和有效性验证。实验结果表明，相较于传统机器学习方法，所提混合模型能够增加源域选择的范围与准确性，它将均方根误差至少降低15%。

Abstract

Short-term load forecasting is crucial for optimizing energy supply and demand balance

which improves the efficiency of power system operation. In regions where historical load data are scarce

traditional machine learning prediction methods face great challenges. To address the short-term load forecasting issue in the context of sparse data

we propose a hybrid prediction model based on temporal convolutional network (TCN) and transfer learning. Firstly

a source domain selection method combining Mahalanobis distance and maximum mean discrepancy is introduced to measure the transferability of the source domain. This method aims to identify a source domain that is highly similar to the target domain in terms of data distribution and feature differences. Secondly

the snow ablation optimizer (SAO) is utilized to optimize the hyperparameters of the TCN model

thereby establishing an SAO-TCN transfer model to enhance overall prediction performance. Finally

the proposed hybrid model is evaluated and validated using real-world datasets. Experimental results demonstrate that in comparison to traditional machine learning methods

the hybrid model is capable of expanding the range of source domain selection and enhancing its accuracy as well

and it reduces the root mean square error by at least 15%.

关键词

Keywords

references

LI Hui, LIU Dong, YAO Danyang. Analysis and reflection on the development of power system towards the goal of carbon emission peak and carbon neutrality[J]. Proceedings of the CSEE, 2021, 41(18): 6245−6259(in Chinese).

HAN Xiaoqing, LI Tingjun, ZHANG Dongxia, et al . New issues and key technologies of new power system planning under double carbon goals[J ] . High Voltage Engineering, 2021, 47(9): 3036−3046(in Chinese).

HAN Fujia, WANG Xiaohui, QIAO Ji, et al . Review on artificial intelligence based load forecasting research for the new-type power system[J ] . Proceedings of the CSEE, 2023, 43(22): 8569−8592(in Chinese).

MA Shaohui. A hybrid deep meta-ensemble networks with application in electric utility industry load forecasting[J]. Information Sciences, 2021, 544: 183−196.

ABDEL-BASSET M, HAWASH H, SALLAM K, et al . STLF-Net: Two-stream deep network for short-term load forecasting in residential buildings[J ] . Journal of King Saud University - Computer and Information Sciences, 2022, 34(7): 4296−4311.

YE R, DAI Q. A relationship-aligned transfer learning algorithm for time series forecasting[J]. Information Sciences, 2022, 593: 17−34.

LIU J, WANG S, WEI N, et al . Natural gas consumption forecasting: A discussion on forecasting history and future challenges[J ] . Journal of Natural Gas Science and Engineering, 2021, 90: 103930.

DENG Daiyu, LI Jian, ZHANG Zhenyuan, et al . Short-term electric load forecasting based on EEMD-GRU-MLR[J ] . Power System Technology, 2020, 44(2): 593−602(in Chinese).

XU Jing, CHI Fujian, GE Leijiao, et al . Short-term combined commercial load forecasting method based on SARIMA-GRNN-SVMM[J ] . Proceedings of the CSU-EPSA, 2020, 32(2): 85−91(in Chinese).

MAI Hongkun, XIAO Jianhong, WU Xichen, et al . Research on ARIMA model parallelization in load prediction based on R language[J ] . Power System Technology, 2015, 39(11): 3216−3220(in Chinese).

MANANDHAR P, RAFIQ H, RODRIGUEZ-UBINAS E. Current status, challenges, and prospects of data-driven urban energy modeling: A review of machine learning methods[J]. Energy Reports, 2023, 9: 2757−2776.

HUANG Feihu, ZHAO Honglei, YI Peiyu, et al . An improved power load forecasting method based on transformer[J ] . Modern Electric Power, 2023, 40(1): 50−58(in Chinese).

ZHAO Pei, Dai Yeming. 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(in Chinese).

TANG Xianlun, CHEN Hongxu, XIONG Deyi, et al . Short-term power load forecasting based on extreme gradient boosting and temporal convolutional network[J ] . High Voltage Engineering, 2022, 48(8): 3059−3067(in Chinese).

ZHANG Lin, LAI Xiangping, ZHONG Shuyong, et al . Electricity load forecasting method based on orthogonal wavelet and long short-term memory neural networks[J ] . Modern Electric Power, 2022, 39(1): 72−79(in Chinese).

WAN A, CHANG Q, AL-BUKHAITI K, et al . Short-term power load forecasting for combined heat and power using CNN-LSTM enhanced by attention mechanism[J ] . Energy, 2023, 282: 128274.

HUA G, SUN Y, LI W. Hybrid load prediction model of 5G base station based on time series decomposition and GRU network with parameter optimization[J]. IET Generation, Transmission Distribution, 2024, 18(8): 1548-1558.

BAI S, KOLTER J Z, KOLTUN V. An empirical evaluation of generic convolutional and recurrent networks for sequence modeling[A]. arXiv, 2018.

LU J, BEHBOOD V, HAO P, et al . Transfer learning using computational intelligence: a survey[J ] . Knowledge-Based Systems, 2015, 80: 14−23.

JUNG S M, PARK S, JUNG S W, et al . Monthly electric load forecasting using transfer learning for smart cities[J ] . Sustainability, 2020, 12(16): 6364.

OUYANG Fulian, WANG Jun, ZHOU Hangxia. Short-term power load forecasting method based on improved hierarchical transfer learning and multi-scale CNN-BiLSTM-Attention[J]. Power System Protection and Control, 2023, 51(2): 132−140(in Chinese).

WEI N, YIN C, YIN L, et al . Short-term load forecasting based on WM algorithm and transfer learning mode l[J ] . Applied Energy, 2024, 353: 122087.

WANG K, YU H, SONG G, et al . Adaptive forecasting of diverse electrical and heating loads in community integrated energy system based on deep transfer learning[J ] . Frontiers in Energy Research, 2022, 10: 1008216.

ZHOU Y, LI X, LIU Y, et al . Transfer learning-based adaptive recursive neural network for short-term non-stationary building heating load prediction[J ] . Journal of Building Engineering, 2023, 76: 107271.

LÓPEZ SANTOS M, DÍAZ GARCÍA S, GARCÍA-SANTIAGO X, et al . Deep learning and transfer learning techniques applied to short-term load forecasting of data-poor buildings in local energy communities[J ] . Energy and Buildings, 2023, 292: 113164.

DE MAESSCHALCK R, JOUAN-RIMBAUD D, MASSART D L. The Mahalanobis distance[J]. Chemometrics and Intelligent Laboratory Systems, 2000, 50(1): 1−18.

DENG L, LIU S. Snow ablation optimizer: A novel metaheuristic technique for numerical optimization and engineering design[J]. Expert Systems With Applications, 2023, 225: 120069.

JIANG Haiyan, LIU Haot ian, SHU Xin, et al . Multi-label transfer learning via maximum mean discrepancy[J ] . Information and Control, 2016, 45(4): 463−470(in Chinese).

XUE H, GUI X, WANG G, et al . Prediction of gas drainage changes from nitrogen replacement: a study of a TCN deep learning model with integrated attention mechanism[J ] . Fuel, 2024, 357: 129797.

LU R, BAI R, LI R, et al . A novel sequence-to-sequence-based deep learning model for multistep load forecasting[J ] . IEEE Transactions on Neural Networks and Learning Systems, 2024: 1-15.

GENG G, HE Y, ZHANG J, et al . Short-term power load forecasting based on PSO-optimized VMD-TCN-attention mechanism[J ] . Energies, 2023, 16(12): 4616.

ZHANG B, WANG S, DENG L, et al . Ship motion attitude prediction model based on IWOA-TCN-Attention[J ] . Ocean Engineering, 2023, 272: 113911.

Views

186

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Load Forecasting in Data-scarce Areas Based on Multi-source Transfer Learning and XGBoost

Research on Multi-energy Coupled Power Load Prediction Based on K-means Clustering and LSTM Model

Resident Load Forecasting Using Federal Learning Based on LSTM Model

An Improved Power Load Forecasting Method Based on Transformer

Wind Power Ultra-short-term Interval Prediction Based on an Adaptive Kernel Density Estimation Algorithm

Related Author

TANG Yikun

TANG Zhiyuan

LUO Tongtong

QIAO Ao

LIU Junyong

SU Yun

BU Fanpeng

GUO Naiwang

Related Institution

School of Logistics, Chengdu University of Information Technology

School of Electrical Engineering, Sichuan University

China Electric Power Research Institute

State Grid Shanghai Municipal Electric Power Company

AI问答

⁰