Effects of Tunnel and Its Ventilation Modes on the Aerodynamic Drag of a Subway Train

Round 1

Reviewer 1 Report

This paper reports an in-situ measurement on the effects of tunnel and its ventilation modes on the aerodynamic drag of a subway train with eight carriages during its routine operation. Some interesting conclusions are obtained through multi-parameter analysis. In my opinion, this article is well organized and I suggest it be published in its current form.

Author Response

Thanks for the review’s comments and positive appraisal. We will go through the manuscript carefully to further improve its quality. All revisions in the manuscript are highlighted with yellow color for easy comparison. 

Reviewer 2 Report

How are the Two modes of tunnel ventilation compared? are the working conditions similar ? and how o check that ?

The introduction needs to be extended.

The data acquisition system is to be described with details.

A photo of the studied zone is to be added and described.

The uncertainties of the experimental measurements are to be evaluated.

Have you checked the repetitivity of the measurements?

 

 

Author Response

1 . How are the Two modes of tunnel ventilation compared? are the working conditions similar? and how to check that?

Reply: The experiments were conducted in the Tsuen Wan (TSW) to Central (CEN) line of Hong Kong Mass Transit Railway Company (MTR) , please see the attached file. The total length of this line is 14.6 Km, among which 13.5 Km is in tunnel. The Train ran from TSW to CEN twice to test the effects of tunnel ventilation mode on the train aerodynamic drag. In the recirculation mode, additional cooled air is pumped into the tunnel through the ventilation shafts at all stations. While, in the free-cooling mode, no cooled air is pumped into the tunnel. Except the difference between tunnel ventilation modes, the other working conditions for the two test runs are exactly the same.

It is worth mentioning that, the test was conducted during the routine operation of TSW to CEN line, implying the possibility that other trains were also running in the tunnel either in front of or behind the test train. Besides, the train might accelerate or decelerate frequently (see Figure 3a in the manuscript) during its entire trip. That is, it is impossible to keep the working conditions of the two test trips exactly the same. Consequently, we only focus on the statistical results rather than those at specific times to uncover the possible effects of the tunnel and its ventilation modes on the aerodynamic forces on the test train.

        We appreciate the reviewer’s comment. The related contents in the manuscript have been revised accordingly to address this point. All revision has been highlighted with yellow color.

* Figure 1. Please see the attached file.

 

2. The introduction needs to be extended.

Reply: Thanks for the reviewer’s suggestion. The introduction has been extended regarding the aerodynamic issues of subway system.

        Please refer to the revised manuscript. All revisions are highlighted with yellow color.

3. The data acquisition system is to be described with details.

Reply: Thanks for the reviewer’s comment.

There are two data acquisition systems located within the first and last carriages, respectively, as shown in figure 2 in the manuscript. All signals, e.g., pressure, hot-wire, train speed, temperature, etc., were simultaneously sampled at a frequency of 2KHz using two identical 16-bit A/D converters. The two data acquisition systems were synchronized for each case to record all the signals simultaneously. The sampling duration for each test is 2000 sec (≈ 33 min), covering the entire trip from TSW to CEN.

Following the reviewer’s suggestion, more detailed description has been added to the revised manuscript. Please refer to revisions in section ‘2 Measurement details’.

 

4. A photo of the studied zone is to be added and described.

Reply: Figure 1 (pleae see the attached file) in this reply shows the line map of Hong Kong MTR. The experiment was conducted in the TSW to CEN line, indicated with red line in the map. The general information about this tested line is given in the manuscript.

Following the reviewer suggestion, we would like to include some of the photos taken in the train parking lot where we had the test train equipped and also in the train during the test runs (please see the attached file). However, these photos can’t be added to the manuscript since the total number of figures are not allowed to exceed 12, as required by the Applied Science.

 

5. The uncertainties of the experimental measurements are to be evaluated.

Reply: Thanks for the reviewer’s suggestion.

        The stagnation pressure at both forward and backward surface were measured directly by the pressure transducers (SMP131 Leeg Co.) with the measurement range of 2 KPa and the accuracy of 0.5% FS. The uncertainty for the mean stagnation pressure was estimated to be ±5%.

The wall shear stress was calculated based on  using the mean velocity measured by the hotwire, which was mounted on the train roof in its viscous sublayer. The hotwires were carefully calibrated in a low-speed wind tunnel with a turbulence intensity lower than 0.5%. The uncertainty in u was estimated about ±4%. The gap Δy between the hot-wire and train roof was adjusted and measured using the mounting facility shown in figure 1 in the manuscript. The relevant dimensions, e.g., diameter of the probe, length of the probe, height of the block, etc., were measured in PolyU laboratory with an uncertainty of no more than ±2%. Thus, the uncertainty in Δy should be smaller than ±5%. The total uncertainty in the wall shear stress  is estimated to be ±9%.

In the present experiment, the test train ran from TSW to CEN twice, all test conditions were similar except the tunnel ventilation modes. Besides, all the equipment used and parameter setting were unchanged in the two test runs. Consequently, the variation in the aerodynamic forces can only be ascribed to the tunnel ventilation modes.

Following the reviewer’s suggestion, the estimation of uncertainty was added to the revised manuscript section 2. Please refer to the revisions highlighted with yellow color.

6. Have you checked the repetitivity of the measurements?

Reply: Thanks for this interesting comment.

Since it was an in-situ test conducted when the subway system operated normally, it is practically impossible to repeat the test under exactly the same test conditions. However, according to the analysis and discission given in ‘Section 3 Measurement results’ and ‘Section 4 Discussion’, the present measurement results conform very well to those reported in the literature, which provides an additional validation on present measurements and conclusions drawn from this work.

Author Response File: Author Response.docx

Reviewer 3 Report

The authors presented an interesting paper on the Effects of tunnel and its ventilation modes on the aerodynamic drag of a subway train. The paper is generally well prepared and can be accepted after minor revision.

An actual photo of the subway train is to be added.

Information on data acquisition systems are to be presented.

An experimental uncertainty study is to be performed.

When comparing the recirculation and the free cooling modes, have you checked the train velocities are similar?

Author Response

Comments from Reviewer 3

The authors presented an interesting paper on the Effects of tunnel and its ventilation modes on the aerodynamic drag of a subway train. The paper is generally well prepared and can be accepted after minor revision.

Reply: Thanks for the reviewer for the comment and evaluation.

 

1 . An actual photo of the subway train is to be added.

Reply：Following the reviewer suggestion, we would like to present some of the photos taken at the train parking lot where we had the test train equipped and also in the train during the test, please see the attached file. However, these photos can’t be added to the manuscript since the total number of figures are not allowed to exceed 12, as required by the Applied Science.

 

2. 2. Information on data acquisition systems are to be presented.

Reply：Thanks for the reviewer’s comment.

There are two data acquisition systems located within the first and last carriages, respectively, as shown in figure 2 in the manuscript. All signals, e.g., pressure, hot-wire, train speed, temperature, etc., were sampled at a frequency of 2KHz using two identical 16-bit A/D converters. The two data acquisition systems were synchronized for simultaneously recording all signals. The sampling duration for each test run is 2000 sec (≈ 33 min), covering the entire trip from TSW to CEN.

Following reviewer’s suggestion, more detailed description has been added to the revised manuscript. Please refer to revisions in section ‘2 Measurement details’. All revision has been highlighted with yellow color.

 

3. 3. An experimental uncertainty study is to be performed.

Reply: Good point.

        The stagnation pressure at both forward and backward surface were measured directly by the pressure transducers (SMP131 Leeg Co.) with the measurement range of 2 KPa and the accuracy of 0.5% FS. The uncertainty for the mean stagnation pressure was approximately ±5%.

The wall shear stress was calculated based on  using the mean velocity measured by the hotwire, which was mounted on the train roof in its viscous sublayer. The hotwires were carefully calibrated in a low-speed wind tunnel with a turbulence intensity lower than 0.5%. The uncertainty in u was estimated to be within ±4%. The gap Δy between the hot-wire and train roof was adjusted and measured using a mounting facility shown in figure 1 in the manuscript. The relevant dimensions, e.g., diameter of the probe, length of the probe, height of the block, etc., were measured in the PolyU laboratory with an uncertainty of no morer than ±2%. Thus, the uncertainty in Δy should be smaller than ±5%. The final uncertainty in the wall shear stress  is estimated to be within ±9%.

In the present experiment, the test train ran from TSW to CEN twice, all test conditions were similar except the tunnel ventilation modes. Besides, all the equipment used and parameter setting were also unchanged in the two test runs. Consequently, the variation in the aerodynamic forces can only be ascribed to the tunnel ventilation modes.

Following reviewer’s suggestion, the estimation of uncertainty was added to the revised manuscript section 2. Please refer to the revisions highlighted with yellow color.

 

4. 4. When comparing the recirculation and the free cooling modes, have you checked the train velocities are similar?

Reply: Thanks for the comments.

Since the test was conducted during the routine operation of TSW to CEN line, implying the possibility that other trains were also running in the tunnel in front of or behind the test train. Besides, the train might accelerate or decelerate frequently (see Figure 3a in the manuscript) during its trips. That is, it is impossible to keep the working conditions of the two test trips exactly the same. Consequently, we only focus on the statistical results rather than those at specific times to uncover the possible effects of the tunnel and its ventilation modes on the aerodynamic forces on the test train.

        As shown in figures 6 and 9 in the manuscript, there are a large amount of data at each velocity, thus we can regress the shear stress and pressure according the train speed and then compare them between the two test runs.

Author Response File: Author Response.docx