Keywords

1 Introduction

As the backbone of urban public transportation, the metro system boasts the advantages of speed, efficiency, and high passenger capacity. However, inevitably, this also brings about significant energy consumption and carbon emission [1]. Ventilation and air-conditioning (VAC) systems dominate the energy consumption of metro stations. Accurate and rational estimation of the air conditioning system's cooling load is crucial for energy-efficient and low-carbon design of the VAC system, as well as ensuring the thermal and humidity environment in the station [2].

Due to the presence of multiple interconnected entrances between metro stations and the external environment, it is challenging to effectively separate the public areas from the outside during actual operation. This inevitably leads to the infiltration of outdoor air into the public areas of the metro station through entrances [3, 4]. To the best knowledge of the authors, there are few studies on the infiltration airflow at the entrances and exits of subway stations. Liu et al. study the natural ventilation of entrances [5]. Zhang et al. use IDA software to establish a one-dimensional model to study the ventilation of entrances [6, 7]. Based on the current design and operational conditions of metro stations, there is a lack of comprehensive understanding and research on the infiltration airflow through entrances. Consequently, the impact of infiltration airflow through entrances is not adequately considered in practical system design. This results in erroneous estimates of the infiltration airflow, ultimately affecting the load estimation and system design of VAC systems in metro stations. To address the research gap, a simulation method utilizing the coupled application of STESS software and Fluent software is proposed for mitigating the misestimation. This study is expected to serve as a valuable reference for the rational estimation of cooling loads and energy-efficient design of VAC systems in metro stations.

2 Methodology

2.1 Station Information

Four typical underground double-layered metro stations in a metro line in North China Plain are surveyed in this study. Figure 1 illustrates the station structure. The station hall is located on the underground first floor, while the platform is situated on the underground second floor. The platform follows a typical island configuration, allowing trains to pass on either side. Platform screen doors (PSDs) are installed between the track area and the platform. For public spaces, the thermal and moisture sources primarily include building envelopes, occupants, equipment, lighting, etc. Due to the thermal inertia of the soil, the heat transfer load from the building envelope is generally considered negligible in the design of underground station VAC systems. The occupant load can be calculated based on passenger flow conditions and the heat generated by individuals. Equipment and lighting are typically calculated using unit area indicators or specific heat emission criteria for individual equipment. The station hall is connected to the external environment through entrances. Permeating airflows induced by these entrances may significantly impact the design of the station's VAC system. The accurate estimation of infiltration cooling loads through entry and exit points is crucial for effective control of the thermal and moisture environment in metro stations and the design of the VAC system.

Fig. 1.
figure 1

Schematic diagram of indoor thermal and humid load composition in an underground metro station with PSD system

Full size image

2.2 Simulation Method

In this study, a simulation method utilizing the coupled application of STESS software and Fluent software is proposed. Besides the well-known CFD software Fluent, STESS is a metro thermal environment simulation and analysis software developed by Tsinghua University. Based on extensive on-site testing at metro stations and theoretical research, it has the capability to predict airflow, air volume, pressure, as well as long-term (monthly) and short-term (hourly) temperature variations of metro systems under various conditions, including different regions, climates, system configurations, passenger densities, train operation schedules, and locations (tunnels and platforms) [8].

In this simulation method, the logical relationship between STESS and Fluent software is illustrated in Fig. 2. A STESS simulation model is developed based on the actual conditions of the surveyed metro line, with the model outputting air pressure variations over time at PSD locations. The air pressure variations serve as inputs for Fluent, providing boundary conditions for simulating the airflow within the metro station. The Fluent simulation model, depicted in Fig. 2, ultimately yields the output of air infiltration volume through entrances. The utilization of the STESS simulation model's output as input for Fluent simulation enhances the fidelity of the simulation boundary, thereby significantly enhancing calculation accuracy.

Fig. 2.
figure 2

Simulation method

Full size image

To verify the grid independence and time step independence of the Fluent simulation, grid and time step refinement were conducted. In the base model, the grid count was set at 0.94 million with a time step of 0.1 s. The grid was refined to 1.23 million, and the time step was reduced to 0.05 s. The comparative simulation results are presented in Fig. 3. The impact of grid and time step refinement on the simulation results is minimal, indicating that the current base model satisfies the independence requirements for both grid and time step.

Fig. 3.
figure 3

Grid Independence and Time Step Independence Testing

Full size image

3 Results and Discussion

3.1 Underestimation in Air Infiltration Through Entrances

To ascertain whether the metro station design phase accurately estimated the cooling load induced by infiltration through entrances, this section involves a comparison between the design drawings and on-site test results. It contrasts the designed infiltration load with the actual load. The on-site test was conducted in August, during which the outdoor parameters closely resembled the design conditions. The comparative results are illustrated in Fig. 4. Examining Fig. 4(a), it is evident that air infiltration through entrances constitutes 1.7%–2.7% in the four metro stations, A, B, C, and D. However, the proportion rises to 15.0%–19.8% in Fig. 4(b). This suggests that the conventional design substantially underestimates the cooling load attributed to infiltration through entrances, posing challenges for effective thermal and humidity environment control in metro stations. In summary, while the cooling load resulting from air infiltration through entrances constitutes a significant portion of the total load, the proportion of this load is substantially underestimated during the schematic design phase.

Fig. 4.
figure 4

A breakdown of cooling load: (a) design cooling load, and (b) actual cooling load.

Full size image

3.2 Underestimation Mitigation with the Proposed Method

The simulation results of air infiltration through entrances will be introduced in this section. For the known four stations, the infiltration volume is influenced by various factors, including train frequency and whether trains in both directions enter or exit the station simultaneously. The simulation results indicate that the air infiltration volume resulting from one pair of trains entering and exiting the station ranges from 940 to 1,930 m3. The train frequency varies throughout the hours, with the air infiltration volume ranging from 1.0 to 1.8 × 104 m3/h. Based on the simulated air infiltration volume, the corresponding cooling load can be estimated.

Fig. 5.
figure 5

Underestimation mitigation by simulation

Full size image

Figure 5 illustrates the mitigation of underestimation using the simulation method under design conditions. The results demonstrate a significant reduction in underestimation when employing the proposed simulation method. The estimation error rate can be decreased from 81.0%–87.5% to 8.3%–50.2% with the application of the proposed simulation method. By using this method, the estimation of station load can be more accurate, and the design of VAC system can be more reasonable and efficient.

4 Conclusions

This study conducted a comparative analysis between the cooling load estimates from design drawings and the actual measurements obtained on-site. The results reveal that, while the cooling load resulting from air infiltration through entrances constitutes a significant portion of the total load, the proportion of this load is substantially underestimated during the schematic design phase.

Moreover, a simulation method is proposed, utilizing a combined application of STESS software and Fluent software. The findings demonstrate that employing this simulation method can effectively mitigate the misestimation, reducing it from 81.0%–87.5% to 8.3%–50.2%.

The simulation method proposed in this paper combines STESS software and Fluent software, which requires high computing power. And the current research object is only underground subway stations. The follow-up research will focus on simplifying the simulation and promoting the application scenarios.