District Heating Daily Gas Consumption Prediction Based on Transformer with Thermal Inertia Prior Position Encoding: A Case Study of Urumqi
1. School of Science, Tianjin University of Commerce, Tianjin 300134, China
2. School of Information Engineering, Tianjin University of Commerce, Tianjin 300134, China
| Abstract: | In recent years, with the increasing attention to ecological protection, the Chinese government has set the carbon neutrality development target. The energy supply of district heating systems in northern cities is gradually shifting from coal to cleaner natural gas. Accurate prediction of natural gas load is critical for energy dispatch planning and demand response. Nevertheless, natural gas load is not only affected by random fluctuations of meteorological factors, but also exhibits complex nonlinearity and significant hysteresis characteristics due to the inherent thermal inertia of heating systems, making it difficult for existing methods to meet the prediction accuracy requirements in practical applications. To address this issue, this paper adopts measured operational data from three heating enterprises in a city and proposes a natural gas load prediction algorithm for district heating systems based on a sequential Transformer model (TI-STM) with thermal inertia position encoding. According to the physical law that recent observations exert stronger impacts while distant ones exert weaker impacts on heating load, an exponentially decaying position encoding is designed. Dimension-wise learnable decay coefficients are adopted to characterize the temporal differences of thermal inertia, replacing the traditional sinusoidal position encoding. Meanwhile, meteorological variables of future time steps are fed into the decoder as known covariates to guide the model to utilize temperature priors. With the multi-head attention mechanism, the model mines complex implicit features within the load sequence and quickly focuses on different key segments of the sequence, mitigating long-range dependency problems and reducing information loss. Comprehensive comparative experiments are carried out against advanced algorithms including SVM, LSTM, MLP, RNN, TCN and TI-STM. The results demonstrate that the proposed model achieves superior prediction accuracy, with mean absolute percentage errors (MAPE) of 1.77%, 2.05% and 2.65%, respectively. |
| Keywords: | Gas Consumption Prediction; Thermal Inertia Position Encoding; Transformer Model; Deep Learning; Hyperparameter Tuning |
| DOI: | 10.57237/j.jest.2026.02.002 |
View:
Downloads:
Received: 30 September 2026
Accepted: 24 October 2026
Published Online: 2 November 2026
References
| [1] | Lin X W, Ma J J, Chen H, Shen F, Ahmad S, Li Z Q. Carbon Emissions Estimation and Spatiotemporal Analysis of China at City Level Based on Multi-Dimensional Data and Machine Learning [J]. Remote Sensing, 2022, 14(13): 3014. https://doi.org/10.3390/rs14133014 |
| [2] | Lin XW, Ma JJ, Chen H, Shen F, Ahmad S, Li ZQ. Carbon Emissions Estimation and Spatiotemporal Analysis of China at City Level Based on Multi-Dimensional Data and Machine Learning. Remote Sens-Basel. 2022; 14(13). 3014. https://doi.org/10.3390/rs14133014 |
| [3] | Zhou J, Zhao YX, Fu TT, Zhou X, Liang GC. Dimension optimization for underground natural gas storage pipeline network coupling injection and production conditions. Energy. 2022; 256. https://doi.org/10.1016/j.energy.2022.124651 |
| [4] | Szega M, Zymelka P, Janda T. Improving the accuracy of electricity and heat production forecasting in a supervision computer system of a selected gas-fired CHP plant operation. Energy. 2022; 239. https://doi.org/10.1016/j.energy.2021.122464 |
| [5] | Lim HS, Kim G. Development of Regression Models considering Time-Lag and Aerosols for Predicting Heating Loads in Buildings. Advances in Civil Engineering. 2018; 2018. https://doi.org/10.1155/2018/4878021 |
| [6] | Ding Y, Zhang Q, Yuan TH, Yang K. Model input selection for building heating load prediction: A case study for an office building in Tianjin. Energy Build. 2018; 159: 254-70. https://doi.org/10.1016/j.enbuild.2017.11.002 |
| [7] | Yuan PL, Lin DM, Wang ZS. Coal consumption prediction model of space heating with feature selection for rural residences in severe cold area in China. Sustainable Cities and Society. 2019; 50. https://doi.org/10.1016/j.scs.2019.101643 |
| [8] | Zhang Q, Tian Z, Ma ZJ, Li GY, Lu YK, Niu JD. Development of the heating load prediction model for the residential building of district heating based on model calibration. Energy. 2020; 205. https://doi.org/10.1016/j.energy.2020.117949 |
| [9] | Geysen D, De Somer O, Johansson C, Brage J, Vanhoudt D. Operational thermal load forecasting in district heating networks using machine learning and expert advice. Energy Build. 2018; 162: 144-53. https://doi.org/10.1016/j.enbuild.2017.12.042 |
| [10] | Xue PN, Jiang Y, Zhou ZG, Chen X, Fang XM, Liu J. Multi-step ahead forecasting of heat load in district heating systems using machine learning algorithms. Energy. 2019; 188. https://doi.org/10.1016/j.energy.2019.116085 |
| [11] | Cai W, Wen XD, Li CE, Shao JJ, Xu JG. Predicting the energy consumption in buildings using the optimized support vector regression model. Energy. 2023; 273. https://doi.org/10.1016/j.energy.2023.127188 |
| [12] | Moayedi H, Bui DT, Dounis A, Lyu Z, Foong LK. Predicting Heating Load in Energy-Efficient Buildings Through Machine Learning Techniques. Applied Sciences-Basel. 2019; 9(20). https://doi.org/10.3390/app9204338 |
| [13] | Wang CD, Yuan JJ, Zhang J, Deng N, Zhou ZH, Gao F. Multi-criteria comprehensive study on predictive algorithm of heating energy consumption of district heating station based on timeseries processing. Energy. 2020; 202. https://doi.org/10.1016/j.energy.2020.117714 |
| [14] | Zhu RJ, Guo WL, Gong XJ. Short-Term Load Forecasting for CCHP Systems Considering the Correlation between Heating, Gas and Electrical Loads Based on Deep Learning. Energies. 2019; 12(17). https://doi.org/10.3390/en12173308 |
| [15] | Xue GX, Qi CY, Li H, Kong XF, Song JC. Heating load prediction based on attention long short term memory: A case study of Xingtai. Energy. 2020; 203. https://doi.org/10.1016/j.energy.2020.117846 |
| [16] | Du J, Zheng JQ, Liang YT, Lu XY, Klemes JJ, Varbanov PS, et al. A hybrid deep learning framework for predicting daily natural gas consumption. Energy. 2022; 257. https://doi.org/10.1016/j.energy.2022.124689 |
| [17] | Liu JY, Wang SX, Wei N, Qiao WB, Li Z, Zeng FH. A clustering-based feature enhancement method for short-term natural gas consumption forecasting. Energy. 2023; 278. https://doi.org/10.1016/j.energy.2023.128022 |
| [18] | Wei N, Yin LH, Li C, Wang W, Qiao WB, Li CJ, et al. Short-term load forecasting using detrend singular spectrum fluctuation analysis. Energy. 2022; 256. https://doi.org/10.1016/j.energy.2022.124722 |
| [19] | 任子豪.基于多气象因子耦合与热惰性补偿的集中供热负荷预测与动态调控策略 [C]//中国城镇供热协会供热技术专业委员会,《区域供热》杂志社有限公司.第九届供热学术年会论文集(2026). 2026: 773-790. https://link.cnki.net/doi/10.26914/c.cnkihy.2026.052343 |
| [20] | 赵晶巍, 夏学彬. 基于热惰性辨识的采暖系统热负荷修正与室温波幅控制研究 [C]. 中国城镇供热协会供热技术专业委员会,《区域供热》杂志社有限公司. 第九届供热学术年会论文集(2026). 2026: 1373-1380. https://link.cnki.net/doi/10.26914/c.cnkihy.2026.052396 |
| [21] | 赵洁, 程鹏月, 陶俊秀.建筑热惰性对供热系统智慧调节的影响分析 [J]. 区域供热, 2024(3): 15-19. https://link.cnki.net/doi/10.16641/j.cnki.cn11-3241/tk.2024.03.003 |
| [22] | Wahl M, Bayer D R, Pruckner M. Benchmarking Transformer and xLSTM for time-series forecasting of heat consumption [C]//2026 IEEE Sustainable Power and Energy Conference, 2026: 1-8. https://api.semanticscholar.org/CorpusID:288258460 |
Join the Editorial Board
We invite active, qualified and high profile scientists and researchers to join as Editorial Board Members.
Join UsBecome a Reviewer
Scholars with a strong interest in reviewing are invited to join the reviewer panel to ensure the quality of the research to be published.
Join Us