Annual hourly tunnel air temperature predictive model for subways

Within sustainable development, the energy consumption of ventilation and air-conditioning (VAC) systems in subway stations is a pressing issue. The subway VAC system’s cooling load is significantly affected by tunnel air temperature. However, there is a lack of a quick and precise method for predicting it. In response to this challenge, this study focuses on the development of a predictive model for the annual hourly tunnel air temperature in subways. In this study, a predictive model, based on observed tunnel air temperature profiles obtained from Subway Thermal Environment Simulation Software (STESS), is proposed. This model combines a daily average temperature model in sinusoidal form with an hourly temperature fluctuation model in piecewise function form. Furthermore, the model has been generalised to suit various working conditions, enhancing its applicability. Model validation shows that the maximum absolute errors of the monthly mean temperature amplitude, the yearly mean tunnel air temperature, and the peak-hour temperature fluctuation are 0.01 °C, 0.50 °C, and 0.04 °C, respectively. This predictive model offers a rapid and accurate solution for estimating annual hourly tunnel air temperature, which can improve cooling load calculations and promote energy-efficient operation within subway VAC systems.