A Deep Learning Approach for Enhanced Real-Time Prediction of Winter Road Surface Temperatures in High-Altitude Mountain Areas

Meng ZHANG; Hua GUO; Jing-yang LI; Li LI; Feng ZHU

doi:10.7307/ptt.v36i5.541

Authors

Meng ZHANG Yunnan Key Laboratory of Digital Communications, Yunnan Yunling Highway Science and Technology Co., Ltd., Yunnan Science Research Institute of Communication Co., Ltd.
Hua GUO Yunnan Key Laboratory of Digital Communications, Yunnan Yunling Highway Science and Technology Co., Ltd., Yunnan Science Research Institute of Communication Co., Ltd.
Jing-yang LI Kunming University of Science and Technology, School of Transportation Engineering
Li LI Chang’an University, School of Electronics & Control Engineering
Feng ZHU Nanyang Technological University, School of Civil and Environmental Engineering

DOI:

https://doi.org/10.7307/ptt.v36i5.541

Keywords:

intelligent transportation system, road surface temperature prediction, ong short-term memory model, combination of feature variables, transferability, mountain highway

Abstract

Low temperatures and icing in winter are significant factors that severely affect highway safety and traffic mobility. To enhance the precision and reliability of real-time winter road surface temperature (RST) prediction, a short-term prediction model is developed that harnesses both feature selection and deep learning. Leveraging meteorological data from a mountain highway in Yunnan, China, the key environmental variables affecting road surface temperature were first extracted using a random forest (RF) model for feature selection. These features were then combined with RST data to construct multiple groups of input variable combinations for the prediction model. A short-term prediction model with a 10-minute update frequency was built using a long short-term memory neural network (LSTM), namely RF-LSTM. The best input variable combination and preset parameters for the prediction model were determined through comparative testing with prevalent machine learning models, and the transferability of the prediction model was verified. The results showed that the best input variable combination for the RF-LSTM prediction model was road surface temperature and air temperature. The model recognised that the short-term RST was affected by long and short-term memory characteristics within a two-hour timeframe. When compared to the RF model, backpropagation (BP) neural network model and the standard LSTM model, the proposed model reduces prediction errors by 59.15%, 31.10% and 20.26%, respectively, while the prediction accuracy is 99.13% within an error margin of ±0.5℃. On the verification dataset, the proposed model maintains its time transferability with an average prediction absolute error of 0.0478. In all, the proposed model not only achieves a higher level of precision in real-time RST predictions but also ensures a more consistent and reliable performance under the challenging conditions of high altitude and mountainous terrain, offering enhanced support for traffic safety and road maintenance decision-making.

References

Theofilatos A, Yannis G. A review of the effect of traffic and weather characteristics on road safety. Accident Analysis & Prevention. 2014;72:244–256. DOI: 10.1016/j.aap.2014.06.017.

Li Y, et al. Probability prediction of pavement surface low temperature in winter based on Bayesian structural time series and neural network. Cold Regions Science and Technology. 2022;194:103434. DOI: 10.1016/j.coldregions.2021.103434.

Sukuvaara T, et al. ITS-Enabled advanced road weather services and infrastructures for vehicle winter testing, professional traffic fleets and future automated driving. In Proceedings of the 2018 ITS World Congress, Copenhagen, Denmark. 2018;17–21.

Crevier L, Delage Y. METRo: A new model for road-condition forecasting in Canada. Journal of Applied Meteorology and Climatology. 2001;40(11):2026–2037. DOI: 10.1175/1520-0450(2001)040.

Coudert O, et al. Optima (Road weather information dedicated to road sections). In Proc. 16th Int. Road Weather Conf. 2012;1–7.

Kangas M, et al. RoadSurf: A modelling system for predicting road weather and road surface conditions. Meteorological Applications. 2015;22(3):544–553. DOI: 10.1002/met.1486.

Kršmanc R, et al. Upgraded METRo model within the METRoSTAT project. In Proc. of the 17th SIRWEC Conference. 2014;30:1–8.

Karsisto V, et al. Comparing the performance of two road weather models in the Netherlands. Weather and Forecasting. 2017;32(3):991–1006. DOI: 10.1175/WAF-D-16-0158.1.

Sokol Z, et al. Ensemble forecasts of road surface temperatures. Atmospheric Research. 2017;187:33–41. DOI: 10.1016/j.atmosres.2016.12.010.

Yin Z, et al. On statistical nowcasting of road surface temperature. Meteorological Applications. 2019;26(1):1–13. DOI: 10.1002/met.1729.

Kršmanc R, et al. Statistical approach for forecasting road surface temperature. Meteorological Applications. 2013;20(4):439–446. DOI: 10.1002/met.1305.

Tang J, et al. Pavement temperature short impending prediction based on ARIMA in winter. Journal of Tongji University (Natural Science). 2017;45(12):1824–1829. DOI: 10.11908/j.issn.0253-374x.2017.12.012.

Xu B, et al. Temperature prediction model of asphalt pavement in cold regions based on an improved BP neural network. Applied Thermal Engineering. 2017;120:568–580. DOI: 10.1016/j.applthermaleng.2017.04.024.

Liu B, et al. Road surface temperature prediction based on gradient extreme learning machine boosting. Computers in Industry, 2018;99:294–302. DOI: 10.1016/j.compind.2018.03.026.

Wang K, et al. Forecasts of road surface temperature in winter based on random forests regression. Meteorological Monthly. 2021;47(1):82–93. DOI: 10.7519/j.issn.1000-0526.2021.01.008.

Qiu X, et al. Prediction of temperature of asphalt pavement surface based on APRIORI-GBDT Algorithm. Journal of Highway and Transportation Research and Development. 2019;36:1–10. DOI: 10.1061/9780784483183.020.

Hochreiter S, Schmidhuber, J. Long short-term memory. Neural Computation. 1997;9(8):1735–1780. DOI: 10.1162/neco.1997.9.8.1735.

Chen X, et al. A safety control method of car-following trajectory planning based on LSTM. Promet-Traffic&Transportation. 2023;35(3):380–394. DOI: 10.7307/ptt.v35i3.118.

Jiang R, et al. Predicting bus travel time with hybrid incomplete data–a deep learning approach. Promet-Traffic&Transportation. 2022;34(5):673–685. DOI: 10.7307/ptt.v34i5.4052.

Tabrizi S, et al. Hourly road pavement surface temperature forecasting using deep learning models. Journal of Hydrology. 2021;603:126877. DOI: 10.1016/j.jhydrol.2021.126877.

Dai B, et al. Hourly road surface temperature LSTM prediction model of expressway in winter. China Safety Science Journal. 2023;33(1):136. DOI: 10.16265/j.cnki.issn1003-3033.2023.01.2215.

Breiman L. Random forests. Machine learning. 2021;45:5–32. DOI: 10.1023/A:1010933404324.

Li H, Guo Y. Estimating factors influencing the capacity of the wide-road-and-narrow-bridge section based on random forest. Promet-Traffic&Transportation. 2023;35(1):1–11. DOI: 10.7307/ptt.v35i1.46.

Hecht-Nielsen R. Theory of the backpropagation neural network. In Neural networks for perception. Academic Press, 1992;65–93.

Gregurić M, et al. Towards the spatial analysis of motorway safety in the connected environment by using explainable deep learning. Knowledge-Based Systems. 2023;269:110523. DOI: 10.1016/j.knosys.2023.110523.

Theofilatos A, Yannis G. A review of the effect of traffic and weather characteristics on road safety. Accident Analysis & Prevention. 2014;72:244–256. DOI: 10.1016/j.aap.2014.06.017.

Dey K, et al. Potential of intelligent transportation systems in mitigating adverse weather impacts on road mobility: A review. IEEE Transactions on Intelligent Transportation Systems. 2015;16(3):1107–1119. DOI: 10.1109/TITS.2014.2371455.

Songchitruksa P, Balke K. Assessing weather, environment, and loop data for real-time freeway incident prediction. Journal of Transportation Research Record. 2006;1959:105–113. DOI: 10.1177/0361198106195900112.