{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,8]],"date-time":"2026-03-08T22:13:21Z","timestamp":1773008001622,"version":"3.50.1"},"reference-count":28,"publisher":"MDPI AG","issue":"8","license":[{"start":{"date-parts":[[2022,8,12]],"date-time":"2022-08-12T00:00:00Z","timestamp":1660262400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Entropy"],"abstract":"<jats:p>A key element for reducing energy consumption and improving thermal comfort on high-speed rail is controlling air-conditioning temperature. Accurate prediction of air supply temperature is aimed at improving control effects. Existing studies of supply air temperature prediction models are interdisciplinary, involving heat transfer science and computer science, where the problem is defined as time-series prediction. However, the model is widely accepted as a complex model that is nonlinear and dynamic. That makes it difficult for existing statistical and deep learning methods, e.g., autoregressive integrated moving average model (ARIMA), convolutional neural network (CNN), and long short-term memory network (LSTM), to fully capture the interaction between these variables and provide accurate prediction results. Recent studies have shown the potential of the Transformer to increase the prediction capacity. This paper offers an improved temporal fusion transformers (TFT) prediction model for supply air temperature in high-speed train carriages to tackle these challenges, with two improvements: (i) Double-convolutional residual encoder structure based on dilated causal convolution; (ii) Spatio-temporal double-gated structure based on Gated Linear Units. Moreover, this study designs a loss function suitable for general long sequence time-series forecast tasks for temperature forecasting. Empirical simulations using a high-speed rail air-conditioning operation dataset at a specific location in China show that the temperature prediction of the two units using the improved TFT model improves the MAPE by 21.70% and 11.73%, respectively the original model. Furthermore, experiments demonstrate that the model effectively outperforms seven popular methods on time series computing tasks, and the attention of the prediction problem in the time dimension is analyzed.<\/jats:p>","DOI":"10.3390\/e24081111","type":"journal-article","created":{"date-parts":[[2022,8,15]],"date-time":"2022-08-15T01:47:21Z","timestamp":1660528041000},"page":"1111","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":12,"title":["An Improved Temporal Fusion Transformers Model for Predicting Supply Air Temperature in High-Speed Railway Carriages"],"prefix":"10.3390","volume":"24","author":[{"given":"Guoce","family":"Feng","sequence":"first","affiliation":[{"name":"School of Artificial Intelligence and Data Science, Hebei University of Technology, Tianjin 300401, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2371-5693","authenticated-orcid":false,"given":"Lei","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence and Data Science, Hebei University of Technology, Tianjin 300401, China"}]},{"given":"Feifan","family":"Ai","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence and Data Science, Hebei University of Technology, Tianjin 300401, China"}]},{"given":"Yirui","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence and Data Science, Hebei University of Technology, Tianjin 300401, China"}]},{"given":"Yupeng","family":"Hou","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence and Data Science, Hebei University of Technology, Tianjin 300401, China"}]}],"member":"1968","published-online":{"date-parts":[[2022,8,12]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Lawrence, M., Bullock, R., and Liu, Z.M. (2019). China\u2019s High-Speed Rail Development, World Bank Publications.","DOI":"10.1596\/978-1-4648-1425-9"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"012089","DOI":"10.1088\/1742-6596\/1986\/1\/012089","article-title":"Comfort evaluation and analysis of high-speed train","volume":"1986","author":"Ding","year":"2021","journal-title":"J. Phys. Conf. Ser."},{"key":"ref_3","first-page":"247","article-title":"A review on artificial intelligence in high-speed rail","volume":"2","author":"Yin","year":"2020","journal-title":"J. Transp. Saf. Secur."},{"key":"ref_4","first-page":"257","article-title":"Passengers Comfort Perception and Demands on Railway Vehicles: A Review","volume":"5","author":"Mendes","year":"2020","journal-title":"KEG"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"318","DOI":"10.1016\/j.applthermaleng.2012.08.041","article-title":"Simplified steady-state modeling for variable speed compressor","volume":"50","author":"Li","year":"2013","journal-title":"Appl. Therm. Eng."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"343","DOI":"10.1109\/59.221222","article-title":"Load forecasting via suboptimal seasonal autoregressive models and iteratively reweighted least squares estimation","volume":"8","author":"Mbamalu","year":"1993","journal-title":"IEEE Trans. Power Syst."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Xu, T., Xu, X., Hu, Y., and Li, X. (2017). An Entropy-Based Approach for Evaluating Travel Time Predictability Based on Vehicle Trajectory Data. Entropy, 19.","DOI":"10.20944\/preprints201703.0213.v1"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"2285","DOI":"10.1016\/S1359-4311(03)00186-8","article-title":"Gray forecast of Diesel engine performance based on wear","volume":"23","author":"Chang","year":"2003","journal-title":"Appl. Therm. Eng."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"234","DOI":"10.1016\/j.jpowsour.2013.01.173","article-title":"Temperature and state-of-charge estimation in ultracapacitors based on extended Kalman filter","volume":"234","author":"Chiang","year":"2013","journal-title":"J. Power Sour."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"191","DOI":"10.1016\/j.apenergy.2015.02.032","article-title":"Recursive wind speed forecasting based on Hammerstein Auto-Regressive model","volume":"145","author":"Maatallah","year":"2015","journal-title":"Appl. Energy"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Lee, S., Kim, C.K., and Kim, D. (2020). Monitoring Volatility Change for Time Series Based on Support Vector Regression. Entropy, 22.","DOI":"10.3390\/e22111312"},{"key":"ref_12","first-page":"821","article-title":"Short-term load forecasting using random forests","volume":"Volume 323","author":"Filev","year":"2015","journal-title":"Proceedings of the 7th IEEE International Conference Intelligent Systems IS\u20192014 (Advances in Intelligent Systems and Computing)"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Gumus, M., and Kiran, M.S. (2017, January 5\u20137). Crude oil price forecasting using XGBoost. Proceedings of the 2017 International Conference on Computer Science and Engineering (UBMK), Antalya, Turkey.","DOI":"10.1109\/UBMK.2017.8093500"},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Hotait, H., Chiementin, X., and Rasolofondraibe, L. (2021). Intelligent Online Monitoring of Rolling Bearing: Diagnosis and Prognosis. Entropy, 23.","DOI":"10.3390\/e23070791"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"2922","DOI":"10.1016\/j.egypro.2019.01.952","article-title":"Short-term load forecasts using LSTM networks","volume":"158","author":"Muzaffar","year":"2019","journal-title":"Energy Procedia"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Jiang, Q., Tang, C., Chen, C., Wang, X., and Huang, Q. (2018, January 1\u20134). Stock price forecast based on LSTM neural network. Proceedings of the Twelfth International Conference on Management Science and Engineering Management, Melbourne, Australia.","DOI":"10.1007\/978-3-319-93351-1_32"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"16453","DOI":"10.1007\/s00500-020-04954-0","article-title":"Temporal convolutional neural (TCN) network for an effective weather forecasting using time-series data from the local weather station","volume":"24","author":"Hewage","year":"2020","journal-title":"Soft Comput."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Yang, Y., and Lu, J. (2022). A Fusion Transformer for Multivariable Time Series Forecasting: The Mooney Viscosity Prediction Case. Entropy, 24.","DOI":"10.3390\/e24040528"},{"key":"ref_19","unstructured":"Li, S., Jin, X., Xuan, Y., Zhou, X., Chen, W., Wang, Y.-X., and Yan, X. (2019). Enhancing the locality and breaking the memory bottleneck of transformer on time series forecasting. arXiv."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"1748","DOI":"10.1016\/j.ijforecast.2021.03.012","article-title":"Temporal Fusion Transformers for interpretable multi-horizon time series forecasting","volume":"37","author":"Lim","year":"2021","journal-title":"Int. J. Forecast."},{"key":"ref_21","unstructured":"Guen, V.L., and Thome, N. (2019). Shape and Time Distortion Loss for Training Deep Time Series Forecasting Models. arXiv."},{"key":"ref_22","unstructured":"Bai, S.J., Kolter, J.Z., and Koltun, V. (2018). An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. arXiv."},{"key":"ref_23","unstructured":"Li, S.J., AbuFarha, Y., Liu, Y., Cheng, M.M., and Gall, J. (2020, January 13\u201319). Ms-tcn++: Multi-stage temporal convolutional network for action segmentation. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA."},{"key":"ref_24","unstructured":"Clevert, D.A., Unterthiner, T., and Hochreiter, S. (2015). Fast and accurate deep network learning by exponential linear units (elus). arXiv."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Zhang, S., and Fang, W. (2021). Multifractal Behaviors of Stock Indices and Their Ability to Improve Forecasting in a Volatility Clustering Period. Entropy, 23.","DOI":"10.3390\/e23081018"},{"key":"ref_26","unstructured":"Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, L., and Polosukhin, I. (2017, January 4\u20139). Attention is All you Need. Proceedings of the Advances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems, Long Beach, CA, USA."},{"key":"ref_27","unstructured":"Oreshkin, B.N., Carpov, D., Chapados, N., and Bengio, Y. (2019, January 6\u20139). N-BEATS: Neural basis expansion analysis for interpretable time series forecasting. Proceedings of the International Conference on Learning Representations, New Orleans, LA, USA."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Challu, C., Olivares, K.G., Oreshkin, B.N., Garza, F., Mergenthaler, M., and Dubrawski, A. (2022). N-hits: Neural hierarchical interpolation for time series forecasting. arXiv.","DOI":"10.1609\/aaai.v37i6.25854"}],"container-title":["Entropy"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1099-4300\/24\/8\/1111\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T00:07:41Z","timestamp":1760141261000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1099-4300\/24\/8\/1111"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,8,12]]},"references-count":28,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2022,8]]}},"alternative-id":["e24081111"],"URL":"https:\/\/doi.org\/10.3390\/e24081111","relation":{},"ISSN":["1099-4300"],"issn-type":[{"value":"1099-4300","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,8,12]]}}}