Study on the Mechanical Behavior of a Large-Segment Fully Prefabricated Subway Station During the Construction Process

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The article presents an interesting case study and offers abundant data.

Although stated (in the abstract and keywords), the soil-structure interaction, in terms of modeling, is not discussed in the text.

Regarding the concrete response description, I suggest using the terms “inelastic” and “nonlinear” instead of “plastic.”

In discussing the structural response, some values of displacements and deflections are given. Please discuss the deflections in the context of the applicable regulations. Also, could such deviations jeopardize the operation of the equipment?

Seismic response is mentioned in the introduction (with reference to other studies), but, apparently, it is not considered in the text. Please mention (in the introduction) that it remains out of the scope of the presented study.

Regarding Figures, probably focusing on the details you consider most important would yield a more reader-friendly presentation. Consider using larger Figures.

 

Specific comments:

1. ... and substantially reducing environmental impacts such as dust and noise pollution, effectively promoting sustainable development in the construction industry. (Lines 38-40) Please define: is this an environmental impact only on the construction site or overall? Can you cite some studies, or is this an intuitive guess?
2. “…lifecycle assessment model based on construction decomposition…” (Line 44). If the bases of life cycle assessment (in the cited work) are to be discussed, more details should be provided.
3. “The model simplifies geometric dimensions…” (Line 169) You are using simplified geometry or reducing dimensions?
4. “…ancillary structures…” (Line 197) Please check!
5. (Lines 209-211) Please check this sentence.
6. Mention in the caption of Fig. 5 that the stress-strain relationships are obtained based on B50010-201 (if this is the case).
7. Do the stress-strain curves for concrete exhibit hardening in the pot-pic response?
8. “Eccentricity κ, compressive strength ratio σb0/σc0, tension-compression yield ratio Kc, and stiffness recovery factors ωc、ωt.” [Lines 273-274] To which model do these parameters refer?
9. Please mention the motivation for choosing an elastic model or a CDP model. (shown in Table 4)
10. Is the model defined by equation (18) original or also based on regulations (as for concrete)?
11. 29 and 30 present experimental data or results obtained by finite-element analysis?
12. “Due to the inherent challenge of prefabricated structures achieving…” (Lines 806-807) “to achieve instead of “achieving”

Author Response

Comments 1:

The article presents an interesting case study and offers abundant data.

Response 1:

We sincerely appreciate the reviewer's acknowledgment of our case study's value and data richness.

Comments 2:

Although stated (in the abstract and keywords), the soil-structure interaction, in terms of modeling, is not discussed in the text.

Response 2:

Thank you for identifying this inconsistency. We agree that the explicit modeling of soil-structure interaction (SSI) was not part of our methodology. Therefore, we have removed all claims about SSI consideration in the modeling description. The revised text in the Abstract (lines 12–14) now states: "The model incorporates a concrete inelastic damage constitutive model and a steel elastic-plastic model, accurately simulates key components..."with no reference to SSI.

Comments 3:

Regarding the concrete response description, I suggest using the terms "inelastic" and "nonlinear" instead of "plastic."

Response 3:

We thank the reviewer for this constructive terminology suggestion. We agree that "inelastic" better describes concrete behavior and have replaced "plastic" with "inelastic" in the constitutive model description. The change is implemented in the Abstract (line 12):"a concrete inelastic damage constitutive model".

Comments 4:

In discussing the structural response, some values of displacements and deflections are given. Please discuss the deflections in the context of the applicable regulations. Also, could such deviations jeopardize the operation of the equipment?

Response 4:

We appreciate this valuable point regarding operational considerations. Our structural analysis specifically addressed construction-phase mechanics, evaluating the reported deflections (e.g., roof settlement: 45 mm, slab deflection: 66.91 mm) against relevant construction tolerance standards. It's important to note that while the applicable structural design codes establish deflection limits primarily for structural integrity, safety, and serviceability under expected loads, they typically do not prescribe explicit deflection criteria specifically geared towards guaranteeing the uninterrupted operation of subsequently installed equipment. Therefore, definitively comparing the observed structural deflections against potential impacts on equipment function is not directly possible within the scope of our analysis, as any potential interference would largely depend on the precise operational tolerances dictated by the specific equipment manufacturers and the detailed interface requirements at their mounting locations, aspects which require separate assessment beyond the core structural regulations.

Comments 5:

Seismic response is mentioned in the introduction (with reference to other studies), but, apparently, it is not considered in the text. Please mention (in the introduction) that it remains out of the scope of the presented study.

Response 5:

We thank the reviewer for highlighting this gap. We have explicitly clarified the scope exclusion of seismic analysis in the Introduction (lines 105–106):"...research gaps persist regarding the mechanical responses of large-segment fully prefabricated subway stations under non-seismic conditions, noting that seismic response analysis remains outside the scope of this investigation."

Comments 6:

Regarding Figures, probably focusing on the details you consider most important would yield a more reader-friendly presentation. Consider using larger Figures.

Response 6:

We agree with this constructive suggestion and have made corresponding adjustments in the text.

Comments 7:

... and substantially reducing environmental impacts such as dust and noise pollution, effectively promoting sustainable development in the construction industry. (Lines 38-40) Please define: is this an environmental impact only on the construction site or overall? Can you cite some studies, or is this an intuitive guess?

Response 7:

Thank you for identifying this omission. We agree that supporting citations are essential for these claims. We have made corresponding modifications in the text, adding reference information on line 40.

Comments 8:

“…lifecycle assessment model based on construction decomposition…” (Line 44). If the bases of life cycle assessment (in the cited work) are to be discussed, more details should be provided.

Response 8:

We sincerely appreciate the reviewer's astute observation. Upon careful reflection, we recognize that describing Chen’s [5] lifecycle assessment methodology in this context—while relevant to prefabricated structures—extends beyond the mechanical scopeof our present study. To maintain rigorous focus on structural mechanics and avoid tangential discussion of sustainability metrics, we have removed the specific phrase "while establishing a lifecycle assessment model based on construction decomposition" from the text (Line 44).

The revised sentence now reads:

"Kunyang Chen [5] designed a low-carbon subway station construction methodology by pioneering novel prefabricated structures with internal bracing systems and large-segment full-preassembly, avoiding unnecessary underground space occupation."

Comments 9:

“The model simplifies geometric dimensions…” (Line 169) You are using simplified geometry or reducing dimensions?

Response 9:

Thank you for seeking clarification on this key modeling aspect. We confirm that the geometric dimensions in our study maintain full-scale 1:1 proportionality to the actual structure (Shenzhen Huaxia Station), and no dimensional reduction or scaling down of the model geometry was applied. However, as noted in the manuscript, the model does incorporate geometric simplifications; specifically, certain complex microscale details inherent in the actual structure, such as tendon anchor pockets (draping points / jacking access holes), were not explicitly modeled in favor of focusing computational resources on capturing the dominant structural behavior at a macro level, while still ensuring the primary structural configuration and member dimensions were accurately represented.

Comments 10:

“…ancillary structures…” (Line 197) Please check!

Response 10:

Thank you for this insightful observation regarding terminology precision. Upon reevaluation, we acknowledge that "ancillary structures" inaccurately implies secondary or non-essential components, whereas the described elements (diaphragm walls, pit bracing, force-transfer keys, and backfill media) constitute critical structural systems enabling construction safety and load transfer.

Comments 11:

(Lines 209-211) Please check this sentence.

Response 11:

Thank you for your careful review of symbol definitions. We have revised lines 209-211 of the manuscript.

Comments 12:

Mention in the caption of Fig. 5 that the stress-strain relationships are obtained based on B50010-201 (if this is the case).

Response 12:

Thank you for ensuring code compliance in our material modeling. We confirm that the title of Figure 5 has been modified.

Comments 13:

Do the stress-strain curves for concrete exhibit hardening in the pot-pic response?

Response 13:

Thank you for your inquiry regarding concrete hardening behavior. In our study of prefabricated subway station structures, the simulated concrete stress-strain curves do not exhibit hardening characteristics.

Comments 14:

“Eccentricity κ, compressive strength ratio σb0/σc0, tension-compression yield ratio Kc, and stiffness recovery factors ωc、ωt.” [Lines 273-274] To which model do these parameters refer?

Response 14:

Thank you for seeking clarification on these constitutive parameters. We confirm that all listed parameters exclusively pertain to the Concrete Damaged Plasticity (CDP) model in ABAQUS.

Comments 15:

Please mention the motivation for choosing an elastic model or a CDP model. (shown in Table 4)

Response 15:

We selected the CDP model over an elastic model due to concrete's quasi-brittle nature: experiments confirm tensile cracks close under compression (hence ω_t=0), while compression microcracks persist under tension (hence ω_c=1). This irreversible damage mechanism renders elastic models incapable of capturing stiffness degradation (e.g., 35% displacement prediction error during backfilling), permanent deformation (5cm diaphragm wall rebound), or stress redistribution (bolt tension >1.1GPa). The CDP model, through plasticity-damage coupling theory, accurately characterizes crack development and bearing capacity degradation in prefabricated structures under transient construction loads, fulfilling GB 50010's mandatory damage simulation requirements for underground structures.

Comments 16:

Is the model defined by equation (18) original or also based on regulations (as for concrete)?

Response 16:

Formula 18 is derived from the standard GB/T 228.1-2021 and has been referenced in lines 290-292 of the text. Thank you for pointing out the issue.

Comments 17:

29 and 30 present experimental data or results obtained by finite-element analysis?

Response 17:

Thank you for your question. The data in Figures 29 and 30 exclusively originate from finite element analysis based on calibrated ABAQUS model outputs, with no inclusion of experimental field measurements. Figure 29 reveals the correlation between diaphragm wall displacements and strut removal during construction stages (maximum increment of 6 cm occurring in Stage 4). Figure 30 quantifies the evolution of contact stresses and bolt tensile forces, demonstrating post-backfill stress divergence reaching 400% differential.

Comments 18:

“Due to the inherent challenge of prefabricated structures achieving…” (Lines 806-807) “to achieve instead of “achieving”

Response 18:

Thank you for your careful guidance. I have made corrections in the article.

Reviewer 2 Report

Comments and Suggestions for Authors

This article presents a CAE analysis of the mechanical behavior, in terms of displacements and stress distribution, of an underground metro station constructed with prefabricated elements across different construction stages. Overall, the article is well written, and the application of the methodology is clearly described. My comments are as follows:

1. In Figure 5(a): This may be a scaling issue; however, concrete has a high capacity to withstand compressive stress. Accordingly, the stress values in the negative region of the graph should be greater than those in the positive region.
2. In Section 3.1.3 it is stated that instruments were installed to monitor displacement at different points of the structure. Further technical information regarding these instruments is required. Specifically, what type of sensors were used, and do they provide real-time measurements?
3. In the same section, it is not clear how the variations in axial forces during assembly were determined. While it is understandable that such variations may occur during the assembly process, the methodology for their determination is not explained. What instruments were employed to measure these forces? Furthermore, the significance of determining these axial forces should be clarified. Were these magnitudes incorporated into the stress simulations?
4. With respect to Figures 7 and 8: The units of force cannot be expressed in MPa. This should be corrected.
5. With respect to Figure 8: Please explain how the structural assembly process was numerically simulated. What parameters were considered to represent the movements generated during assembly? Providing this explanation would aid in understanding the force results at points 3, 7, and 10, as shown in Figure 8. Additionally, what is the rationale for presenting results only for those specific points?
6. Why were dynamic loads not taken into account in the simulations?
7. If one of the objectives of the study is to propose alternative construction methods, why was the stress distribution of joint configurations other than the mortise–tenon not examined? Variations in the geometry of the mortise–tenon joint could be considered in the simulations, and their corresponding stress distributions compared.
8. In Section 3.1 it is stated that CAE analyses will contribute to the development of optimization methodologies for the construction of large-segment fully prefabricated subway station structures. However, the manuscript does not specify what these new optimal construction methodologies would entail. A concise summary of the proposed alternative construction (assembly) methods for prefabricated elements in this application should be included.
9. A comparative analysis of the advantages and disadvantages of prefabricated structures versus cast-in-situ structures is also recommended.
10. Finally, and importantly, the quality of the figures throughout the manuscript should be improved. For instance, in the stress color bars, the magnitude values are not clearly distinguishable.

Author Response

Comments 1:

In Figure 5(a): This may be a scaling issue; however, concrete has a high capacity to withstand compressive stress. Accordingly, the stress values in the negative region of the graph should be greater than those in the positive region.

Response 1:

The suggestion you provided is very important. The original image may be misleading to readers. Additional explanations have been provided in the image title. Thank you again for your highly constructive feedback.

Comments 2:

In Section 3.1.3 it is stated that instruments were installed to monitor displacement at different points of the structure. Further technical information regarding these instruments is required. Specifically, what type of sensors were used, and do they provide real-time measurements?

Response 2:

Thank you for your valuable guidance. We have added relevant information about the instrument and provided a detailed explanation at 320-327. The monitoring plan uses steel bar gauges, concrete strain gauges, force transmission key force gauges, strain gauges, wireless acquisition devices, and side wall and roof force transmission key sensor acquisition instruments as shown in Figure 6 (c).

Comments 3:

In the same section, it is not clear how the variations in axial forces during assembly were determined. While it is understandable that such variations may occur during the assembly process, the methodology for their determination is not explained. What instruments were employed to measure these forces? Furthermore, the significance of determining these axial forces should be clarified. Were these magnitudes incorporated into the stress simulations?

Response 3:

Thank you for your question. Axial force is monitored using a steel bar gauge, and a portion of the monitoring data will be used to verify the validity of the numerical model. The comparison results are shown in Figure 8.

Comments 4:

With respect to Figures 7 and 8: The units of force cannot be expressed in MPa. This should be corrected.

Response 4:

Thank you for pointing out your valuable opinion. We agree with your statement because axial force is generally expressed in N. However, we have processed the data and considered its force on the plane of the steel bar axis. Because the sensor automatically completes the unit conversion based on its monitoring principle in the equipment, the output result is directly the surface force on the circular section of the steel bar, expressed as Pa. If you have any further questions, please let us know. Thank you again for your valuable opinion.

Comments 5:

With respect to Figure 8: Please explain how the structural assembly process was numerically simulated. What parameters were considered to represent the movements generated during assembly? Providing this explanation would aid in understanding the force results at points 3, 7, and 10, as shown in Figure 8. Additionally, what is the rationale for presenting results only for those specific points?

Response 5:

Thank you for pointing out the issue. In response to the numerical simulation in Figure 8, we have adopted a seven stage sequential activation strategy to accurately reproduce the structural assembly process: starting from the activation of the foundation pit support system, we sequentially add prefabricated bottom plate blocks and apply prestress (through equivalent simulation of thermal expansion coefficient), install middle columns and longitudinal beams (using bolt connections), lift side walls (considering circumferential steel reinforcement constraints), lay middle plates (simulating cow leg contact and bolt stress), assemble top plate arch blocks (introducing prestressed steel bundles), and finally backfill loading (simulating the self weight of the covering soil). The entire process considers key parameters such as component contact friction (μ=0.6), concrete damage evolution (CDP model), and prestress loss. Selecting monitoring points 3 (roof arch foot), 7 (middle of side wall), and 10 (mid span of bottom plate) as they are located on the critical path of structural force transmission, together form the core indicator system for evaluating the overall mechanical behavior.

Comments 6:

Why were dynamic loads not taken into account in the simulations?

Response 6:

Thank you very much for pointing out the issue. This study explicitly excludes seismic response analysis from the scope of research and focuses on non seismic static behavior simulation during the construction process. The aim is to fill the gap in the mechanical response of large fully prefabricated subway stations under static loads such as self weight and overburden pressure during the construction phase, ensuring that the analysis focuses on the stability assessment of the structural assembly process, rather than the effects of dynamic loads such as earthquakes or vibrations.

Comments 7:

If one of the objectives of the study is to propose alternative construction methods, why was the stress distribution of joint configurations other than the mortise–tenon not examined? Variations in the geometry of the mortise–tenon joint could be considered in the simulations, and their corresponding stress distributions compared.

Response 7:

Thank you for your highly professional question. This is the core goal of our prefabricated engineering research, and your opinion is very important. This study focuses on the mortise and tenon joint structure actually used at Shenzhen Huaxia Station (as shown in Figure 3), and its geometric parameters have been verified through engineering and standardized design. We have published a parametric study on other joint forms, such as the influence of geometric variations in force transmission keys on stress performance, in the journal Sustainability (Qinglou Li, Yuanzhuo Li et al., 2024). This paper focuses on simulating the system mechanical behavior throughout the construction process, ensuring the depth and completeness of the core objective - the stability assessment of prefabricated assembly processes, and avoiding excessive expansion of research scope.

Comments 8，9:

In Section 3.1 it is stated that CAE analyses will contribute to the development of optimization methodologies for the construction of large-segment fully prefabricated subway station structures. However, the manuscript does not specify what these new optimal construction methodologies would entail. A concise summary of the proposed alternative construction (assembly) methods for prefabricated elements in this application should be included. A comparative analysis of the advantages and disadvantages of prefabricated structures versus cast-in-situ structures is also recommended.

Response 8，9:

Thank you for your valuable feedback. Your guidance is very professional. We should indeed provide a comprehensive summary of the large block fully assembled subway station. We have supplemented its advantages in the conclusion of the article, as shown in lines 830-836 of the paper. Thank you again for your feedback!

Comments 10:

Finally, and importantly, the quality of the figures throughout the manuscript should be improved. For instance, in the stress color bars, the magnitude values are not clearly distinguishable.

Response 10:

Thank you for your professional advice. We have made adjustments to the images that require highlighting details to make them easier to observe. Thank you again for your feedback.

Reviewer 3 Report

Comments and Suggestions for Authors

Review on the article «Study on Mechanical Behavior of Large-Segment Fully Prefabricated Subway Station During Construction Process» by Tan Zhongsheng, Li Yuanzhuo, Fan Xiaomin, Wang Jian

The development of engineering methods for studying sections of critical public structures is of great importance for modern structural mechanics, since it is their stability and reliability that determine the durability of infrastructure facilities. The article is devoted to this relevant problem. After conducting a literature review, the authors considered the applied problem of studying fragments of Shenzhen metro structures.

Having familiarized themselves with the work, the reviewer came to the conclusion that the applied orientation of the results is its main advantage.

However, there are a number of comments to the work that should be emphasized:

1. References. Almost the entire list of references is to Chinese authors. Is this problem not relevant in other countries? The reviewer understands that the publication concerns metro structures in China, but, probably, similar problems are also characteristic of other countries.
2. The use of relations (1) – (22) requires references to the relevant literary sources.
3. The reviewer believes that the article should include the parameters of the devices on which the calculations were performed and indicate the average time of one numerical experiment. This is important, since the article is devoted specifically to the numerical modeling of engineering structures.
4. The components presented in Fig. 4 and Fig. 11 - 28 are quite small and cannot be read. This should be changed.

In general, the comments made by the reviewer are aimed at improving the informativeness and perception of the work. The reviewer believes that this work is interesting and useful for the readers of the journal and can be published after taking into account the corrections indicated above.

Author Response

Comments 1:

References. Almost the entire list of references is to Chinese authors. Is this problem not relevant in other countries? The reviewer understands that the publication concerns metro structures in China, but, probably, similar problems are also characteristic of other countries.

Response 1:

We acknowledge the reviewer's observation regarding reference distribution.Thank you for your valuable feedback. However, due to the limited number of similar engineering research articles abroad, Chinese authors make up a significant portion of the content related to this article. Nevertheless, we still consider the need to cite articles from other countries in lines 40-47 of the article.

Comments 2:

The use of relations (1) – (22) requires references to the relevant literary sources.

Response 2:

Thank you for your valuable suggestions on our research. We have cited the formulas in the article. As the majority of the formulas are derived from the two standards cited in the article, we have cited them in lines 220-22, 295-297 for the sake of conciseness.

Comments 3:

The reviewer believes that the article should include the parameters of the devices on which the calculations were performed and indicate the average time of one numerical experiment. This is important, since the article is devoted specifically to the numerical modeling of engineering structures.

Response 3:

We sincerely appreciate your insightful advice. All calculations were performed on our laboratory's dedicated small-scale server, with a total simulation time of approximately 5 calendar days (including necessary debugging). We will add detailed device specifications and average single-experiment time in the revised manuscript. The corresponding content has been added in lines 209-211 of the text.

Comments 4:

The components presented in Fig. 4 and Fig. 11 - 28 are quite small and cannot be read. This should be changed.

Response 4:

Thank you for your valuable suggestions on our research. We have enlarged the images in the article in a targeted manner to ensure the effectiveness of the presentation.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors responded to most of the comments I formulated in the previous review round. Some additional suggestions (the numbering of the comments follows the response letter provided by the authors).

 

Regarding comment 12:

Suggestion for the caption of Figure 5:

Figure 5. Stress-strain relationships obtained by using B50010-201

 

Regarding comment 15:

I am précising further specifying my question. In Table 4, some elements (for example, “the prefabricated component”) are modeled by using CDP, while others (for example, the “Underground continuous wall”) are modeled by using an elastic model.

Please discuss the choice of CDP/elastic model in the different cases listed in Table 4 (for example, the expected stress and strain state in a given element or other reasons).

 

Regarding comment 17:

Please mention this in the captions of Figures 29 and 30 (the manuscript also contains experimental results).

Author Response

Comments 1:

Suggestion for the caption of Figure 5:

Figure 5. Stress-strain relationships obtained by using B50010-201

Response 1:

Thank you for your suggestion. We will adopt your suggestion and make modifications to the title in Figure 5.

Comments 2:

I am précising further specifying my question. In Table 4, some elements (for example, “the prefabricated component”) are modeled by using CDP, while others (for example, the “Underground continuous wall”) are modeled by using an elastic model.

Please discuss the choice of CDP/elastic model in the different cases listed in Table 4 (for example, the expected stress and strain state in a given element or other reasons).

Response 2:

Thank you for your insightful comments. The CDP model was selected for the prefabricated component and concrete support to accurately capture their nonlinear behavior, crack development, and stiffness degradation under significant transient construction loads and stress redistribution, as indicated by their full set of damage parameters (e.g., f_c,r, ε_c,r, α_c). Conversely, the underground continuous wall and fertilizer tank backfilling, modeled as elastic, are subjected to relatively stable stress states (e.g., predominant hydrostatic pressure or minor loads) where nonlinear damage effects are negligible, thus simplifying the analysis without compromising accuracy, consistent with their omitted damage parameters in Table 4. We sincerely appreciate your valuable feedback.

Comments 3:

Please mention this in the captions of Figures 29 and 30 (the manuscript also contains experimental results).

Response 3:

Thank you for your valuable feedback. We have decided to annotate the captions of Figures 29 and 30. Once again, we appreciate your valuable input.

Reviewer 3 Report

Comments and Suggestions for Authors

My comments have been taken into account. I propose to accept.

Author Response

Thank you for providing valuable reference suggestions for this article, which has made great progress in our work. Once again, thank you for your guidance on this article and for your recognition.