Multi-Model Intelligent Prediction of Rock Integrity in Tunnels Based on Geological Differences of Ground-Penetrating Radar Exploration Workfaces

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study proposes a model for intelligently identifying tunnel surrounding rock integrity using ground-penetrating radar (GPR) detection results and applies the model in 10 tunnels.  However, the manuscript lacks sufficient logical consistency and data rigor.  Ground-penetrating radar acquires reflected electromagnetic waves from the rock mass and structural surfaces within a certain range in front of the antenna.  Establishing relationships between rock masses of different integrity levels and GPR signal characteristics is the foundation for identifying rock integrity.  

The biggest flaw in this manuscript is that it does not provide typical GPR data for the five surrounding rock integrity levels (Complete, Fairly Complete, Fairly Broken, Broken, Extremely Broken) shown in Figure 2.  Reviewers are thus unable to determine whether there is sufficient valid raw data to support the construction of the recognition model.  In fact, the signal characteristics of GPR waves propagating in rock are also related to factors such as rock type and signal attenuation properties, which this manuscript fails to consider.

Additionally, the article has shortcomings in its expression and writing standards.  For example, line 118 of the main text refers to Figure 1 as the working principle diagram of ground-penetrating radar, but Figure 1 does not depict the propagation of electromagnetic waves in rock and structural fractures.  Furthermore, line 274 contains corrupted figure citations.

In summary, despite technical novelty, the model construction lacks comprehensiveness and rigor, features insufficient data analysis, and exhibits limited depth.  The work overemphasizes algorithmic descriptions at the expense of broader geological insight.  Consequently, publication in this journal is not advised.

Author Response

First of all, the authors would like to thank the reviewer for your extremely detailed review of the paper. Your valuable comments and suggestions helped us improving its quality and presentation. Below, please find our reply to your comments. We have addressed the comments very carefully and hope that the revision is now up to the standards of Infrastructures.

Comment 1. This study proposes a model for intelligently identifying tunnel surrounding rock integrity using ground-penetrating radar (GPR) detection results and applies the model in 10 tunnels. However, the manuscript lacks sufficient logical consistency and data rigor. Ground-penetrating radar acquires reflected electromagnetic waves from the rock mass and structural surfaces within a certain range in front of the antenna. Establishing relationships between rock masses of different integrity levels and GPR signal characteristics is the foundation for identifying rock integrity. 

The biggest flaw in this manuscript is that it does not provide typical GPR data for the five surrounding rock integrity levels (Complete, Fairly Complete, Fairly Broken, Broken, Extremely Broken) shown in Figure 2.  Reviewers are thus unable to determine whether there is sufficient valid raw data to support the construction of the recognition model. In fact, the signal characteristics of GPR waves propagating in rock are also related to factors such as rock type and signal attenuation properties, which this manuscript fails to consider.

Reply:

We sincerely appreciate your valuable feedback on the GPR detection schematic in Figure 1. We fully recognize the issues of small image size and unclear expression, which have hindered readers from effectively grasping the layout and working procedures of the GPR. In response to the aforementioned issues, we have made modifications to this diagram.

In 4.1. Dataset preparation ……The dataset containing 694 samples is constructed, covering three rock integrity grades—Fairly Complete, Fairly Broken, and Broken. Training data come from 34 tunnel sites, with 90 batches of detection data, and test data come from 10 tunnel sites, with 29 batches of detection data. Test data comes from 10 tunnel sites, with 29 batches of detection data.

Comment2. Additionally, the article has shortcomings in its expression and writing standards.  For example, line 118 of the main text refers to Figure 1 as the working principle diagram of ground-penetrating radar, but Figure 1 does not depict the propagation of electromagnetic waves in rock and structural fractures.  Furthermore, line 274 contains corrupted figure citations.

Reply:

Thank you for drawing our attention to the shortcomings in Figure 1. We have replaced the original image with a new schematic that explicitly illustrates the emission of high-frequency electromagnetic waves, their propagation through intact rock and fractured zones, and the consequent reflections captured by the receiving antenna, thereby giving a clearer visual explanation of the GPR working principle.

In addition, the corrupted figure reference that appeared at line 274 of the previous version has been corrected, and we have conducted a full pass through the manuscript to confirm that all figure citations are now accurate and consistent.

In 2.2. the interpretation mechanism of GPR Figure 1. GPR detection schematic

Comment3. In summary, despite technical novelty, the model construction lacks comprehensiveness and rigor, features insufficient data analysis, and exhibits limited depth.  The work overemphasizes algorithmic descriptions at the expense of broader geological insight.  Consequently, publication in this journal is not advised.

Reply:

Once again, we sincerely thank the reviewer for these constructive suggestions. We have revised the manuscript accordingly and believe the changes fully address the concerns raised, further improving the clarity and rigor of the work. We hope the updated version now meets the journal’s requirements and look forward to your favorable consideration.

Finally, the following acknowledgement is added to express our sincere gratitude to you.

 

In Acknowledgement

……

    Moreover, the authors would like to thank the editors and anonymous reviewers for their numerous detailed and inspiring suggestions and comments that helped improving the quality and readability of this paper.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This is a study titled “Multi-model Intelligent Prediction of Rock Integrity in Tunnels Based on Geological Differences of Ground Penetrating Radar Exploration Workfaces”, a multi-model intelligent prediction method for tunnel rock integrity based on geological differences of GPR exploration workfaces was proposed in the paper. The research is meaningful and helpful for practice, but it requires significant modifications in expression. The main comments are as follows:

1. Regarding the GPR detection schematic in Figure 1, firstly, the image size is too small to see clearly, and secondly, the expression is not clear. Readers cannot obtain the relevant layout and working procedures of GPR through the schematic diagram. Please optimize and modify the diagram, and finally add the layout and process of relevant instruments.
2. In the introduction chapter, it is not necessary to introduce what each chapter does, just like in Chinese articles. You just need to logically explain the reasons, methods, expected results, etc. for what you should do.
3. Regarding the quantitative standards for rock integrity, there are some regulations in China that provide detailed explanations and classifications. It is suggested that authors can further refer to them.
4. The font size of the annotations on the coordinate axes in Figures 3 to 6 is too small, making it difficult for people to see clearly. Another issue is why the horizontal axis can be written horizontally, but why is it changed to an oblique placement? These figures lack relevant legends and main data, and the authors need to optimize and supplement them.
5. Authors should elaborate on the uniqueness of their proposed methods and why they are accurate. And in the discussion section, compare and analyze the results of one's own method with those of others, and further evaluate the advantages and disadvantages of one's own method.
6. What is the relationship between Conclusion and Main Contributions of This Article? In my opinion, there is no need to make a distinction. In the conclusion section, the innovative points and important conclusions or contributions of the article can be elaborated.
7. Some details of the article need to be processed and optimized, and it seems like it was written by a novice. In addition, the article has fewer figures and should be described with richer and more figures. The references in the article are also too few. As far as I know, there are many relevant literature on the integrity of tunnel rocks, and the authors should greatly enrich them.

 

Comments on the Quality of English Language

Read the entire text to verify grammar and expression

Author Response

First of all, the authors would like to thank the reviewer for your extremely detailed review of the paper. Your valuable comments and suggestions helped us improving its quality and presentation. Below, please find our reply to your comments. We have addressed the comments very carefully and hope that the revision is now up to the standards of Infrastructures.

Comment 1. Regarding the GPR detection schematic in Figure 1, firstly, the image size is too small to see clearly, and secondly, the expression is not clear. Readers cannot obtain the relevant layout and working procedures of GPR through the schematic diagram. Please optimize and modify the diagram, and finally add the layout and process of relevant instruments.

Reply:

We sincerely appreciate your valuable feedback on the GPR detection schematic in Figure 1. We fully recognize the issues of small image size and unclear expression, which have hindered readers from effectively grasping the layout and working procedures of the GPR. In response to the aforementioned issues, we have made modifications to this diagram.

In 2.2. The interpretation mechanism of GPR Figure 1. GPR detection schematic

Comment2. In the introduction chapter, it is not necessary to introduce what each chapter does, just like in Chinese articles. You just need to logically explain the reasons, methods, expected results, etc. for what you should do.

Reply:

Thank you for your valuable suggestion regarding the introduction chapter. We have fully taken it into account and made the necessary adjustments. In the revised version, instead of listing the content of each chapter, we have reorganized the introduction to focus on the logical presentation of the research motivation, methodology, and expected outcomes.

In 1. Introduction ……, which can effectively improve the prediction accuracy. To achieve this, we first analyze the structural and physical characteristics of fractured surrounding rock and elucidate the working principles and interpretation mechanisms of GPR. Subsequently, we develop a specific implementation process for our proposed multi-model intelligent prediction method, which takes into account the geological differences at the exploration workface. This method is then applied to actual engineering sites, where we utilize it to forecast and verify the conditions ahead of the tunnel face. The results obtained from these applications are thoroughly discussed to evaluate the method’s effectiveness. Finally, we summarize the key contributions of this research, highlighting its potential for enhancing the safety and efficiency of tunnel construction through more accurate geological predictions.

Comment3. Regarding the quantitative standards for rock integrity, there are some regulations in China that provide detailed explanations and classifications. It is suggested that authors can further refer to them.

Reply:

Thank you for your suggestion. We have revised reference [21] to the well - established and widely - used published book, the "Engineering Rock Mass Classification Standard". This standard comprehensively guides rock mass classification for various rock engineering projects. Additionally, we've added detailed introductions to RMR and Q - system along with relevant citations. These adjustments enhance the clarity and credibility of the section.

In 2.1. Rock integrity characteristics and fractured rock structure To quantitatively evaluate rock mass quality and guide tunnel support design, several internationally recognized classification systems have been developed. Among them, the Rock Mass Rating (RMR) system and the Q-system are two widely used approaches [23]. The RMR system evaluates rock mass based on parameters including uniaxial compressive strength, rock quality designation (RQD), joint spacing, joint condition, groundwater inflow, and orientation adjustment [24]. The Q-system, on the other hand, integrates RQD, number of joint sets, joint roughness, joint alteration, stress reduction factor, and groundwater conditions to provide a comprehensive index of rock mass quality [25]. These systems are essential tools for engineers to assess the stability and support requirements of rock masses in various geological settings. Despite the widespread use of RMR and Q-systems, different projects may require specific criteria tailored to local geological conditions. In this study, we adopt the engineering rock mass classification standard as defined by relevant regulations in China [26]. According to this standard, rock integrity is qualitatively classified into five grades: complete, fairly complete, slightly broken, broken, and extremely broken. This classification is based on key indicators such as joint characteristics (including spacing, continuity, and filling materials), the types of dominant structural planes, and their degree of interlocking. These classifications reflect the degree of fracturing and continuity within the rock mass, which directly influence its mechanical behavior and stability. This five-grade classification provides a practical framework for assessing the integrity of rock ahead of tunnel excavation, ensuring appropriate measures are taken to mitigate potential risks during construction.

Comment4. The font size of the annotations on the coordinate axes in Figures 3 to 6 is too small, making it difficult for people to see clearly. Another issue is why the horizontal axis can be written horizontally, but why is it changed to an oblique placement? These figures lack relevant legends and main data, and the authors need to optimize and supplement them.

Reply:

We sincerely appreciate your meticulous review and constructive feedback on Figures 3 to 6. We fully acknowledge the issues regarding the font size of axis annotations, the orientation of the horizontal axis, and the lack of legends and key data, and have taken immediate steps to address them comprehensively.

In 3.1.1. Obtaining relative amplitude matrix of GPR Figure 3. The data distribution before division Figure 4. The working face is fairly complete Figure 5. The working face is slightly broken Figure 5. The working face is broken

 

Comment5. Authors should elaborate on the uniqueness of their proposed methods and why they are accurate. And in the discussion section, compare and analyze the results of one's own method with those of others, and further evaluate the advantages and disadvantages of one's own method. What is the relationship between Conclusion and Main Contributions of This Article? In my opinion, there is no need to make a distinction. In the conclusion section, the innovative points and important conclusions or contributions of the article can be elaborated.

Reply:

We truly appreciate your perspective regarding the relationship between the "Conclusion" and "Main Contributions" sections. We fully understand your suggestion that the innovative aspects, significant conclusions, and contributions could be more cohesively presented within a unified conclusion section.

In 5. Conclusion Aiming at the problem that the response characteristics of geological radar to adverse geology are complicated and the characteristics of data distribution are not obvious, this study proposes a multi-model intelligent method for predicting rock integrity in tunnels based on geological differences of GPR exploration workfaces. This method achieves innovative breakthroughs and practical contributions in the following aspects: l   Proposes a method for extracting structured feature matrices from unstructured GPR detection data, addressing the challenge of converting raw radar data into analyzable formats for intelligent prediction. l   Introduces an index to measure amplitude anomaly fluctuations, namely the sum of row variances, which provides a quantitative basis for identifying abnormal data in GPR signals. l   \item Considers the geological differences of GPR exploration workfaces to establish multi-models for predicting rock integrity, improving the adaptability of the prediction method to complex and variable tunnel geological conditions. l   \item Validates the effectiveness of the proposed method through testing in ten real tunnels and comparison with two alternative methods. The method achieves an accuracy of 95.33\% in the constructed dataset, with precision rates above 90\% for the prediction of fairly complete, slightly broken, and broken rock categories, demonstrating its practical application value. Despite these achievements, limitations exist: the dataset used in this study only covers three conditions of workfaces, and the results for fractured workface categories lack sufficient diversity, which may reduce the persuasiveness of the findings. Future research will focus on expanding the dataset to include more geological conditions and workface scenarios, thereby enhancing the method's robustness and generalizability. Overall, this study provides a significant advancement in the application of GPR data for tunnel rock integrity prediction, laying a foundation for more accurate and reliable geological risk assessment in tunnel construction.

Comment6. Some details of the article need to be processed and optimized, and it seems like it was written by a novice. In addition, the article has fewer figures and should be described with richer and more figures. The references in the article are also too few. As far as I know, there are many relevant literature on the integrity of tunnel rocks, and the authors should greatly enrich them.

Reply:

We have substantially expanded the bibliography by adding several additional publications that focus specifically on tunnel-rock integrity evaluation and recent advances in GPR interpretation. These references include both seminal papers and the latest studies (2022–2025), providing a more comprehensive scholarly context.

Comment7. Read the entire text to verify grammar and expression

Reply:

Following your comment, the grammar and spelling have been overhauled by using a well-known tool “Grammarly”. The revisions are not shown here since there are many of them. Please refer to the manuscript.

Finally, the following acknowledgement is added to express our sincere gratitude to you.

 

In Acknowledgement

……

    Moreover, the authors would like to thank the editors and anonymous reviewers for their numerous detailed and inspiring suggestions and comments that helped improving the quality and readability of this paper.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

I have gone through MS entitled "Multi-model Intelligent Prediction of Rock Integrity in Tunnels Based on Geological Differences of Ground Penetrating Radar Exploration Workfaces", where the authors used GPR to extract the tunnel workface geological data. A statistic is proposed to identify abnormal data, and filtering rules are formulated to eliminate these anomalies. Then, considering the difference of geological conditions of the GPR exploration workface, multiple models are established with different degrees of fragmentation of the exploration workface. Finally, the validity of the multi-model prediction method is proved by practical engineering verification. 

There are few minor revision require which are given below.

"During the construction of such tunnels, fractured rock is often encountered, which is less stable and prone to accidents such as falling blocks and collapses. " Authors should use some references like https://doi.org/10.1007/s43939-024-00144-z, doi:10.12691/jcd-1-1-3,

"The rest of the paper is organized as follows. Chapter 2 introduces the structural 82
and physical characteristics of the fractured surrounding rock and describes the working....." Authors should rephrase these sentences and rewrite them as a research article, not like a book or thesis.

Section 2.1 Rock mass classification is crucial for designing tunnel support. It includes rock mass strength, geological discontinuities conditions, orientation and ground water status. Common classification systems include RMR (Rock Mass Rating) and the Q-system, which use different parameters to classify the rock mass

Authors should use the above mentioned classification of Rock Mass instead of using some report as cited reference 21. 

Figure 2 authors mentioned used geological informations in data pre processing, please clarify what are geological informations used?

 

Author Response

First of all, the authors would like to thank the reviewer for your extremely detailed review of the paper. Your valuable comments and suggestions helped us improving its quality and presentation. Below, please find our reply to your comments. We have addressed the comments very carefully and hope that the revision is now up to the standards of Infrastructures.

Comment1."During the construction of such tunnels, fractured rock is often encountered, which is less stable and prone to accidents such as falling blocks and collapses. " Authors should use some references like https://doi.org/10.1007/s43939-024-00144-z, doi:10.12691/jcd-1-1-3.

Reply:

Thank you for your suggestion. We have incorporated the relevant literature citations as recommended to support the statement about the stability issues of fractured rock in tunnel construction, thereby enhancing the academic rigor of this part of the content.

In 1. Introduction ……During the construction of such tunnels, fractured rock is often encountered, which is less stable and prone to accidents such as falling blocks and collapses [2,3].

Comment2."The rest of the paper is organized as follows. Chapter 2 introduces the structural 82 and physical characteristics of the fractured surrounding rock and describes the working....." Authors should rephrase these sentences and rewrite them as a research article, not like a book or thesis.

Reply:

Thank you for your valuable suggestion on refining the structure description. We agree that the original wording had a overly thesis-like tone, and we have revised it to align with the concise style of research articles.

In 1. Introduction ……, which can effectively improve the prediction accuracy. To achieve this, we first analyze the structural and physical characteristics of fractured surrounding rock and elucidate the working principles and interpretation mechanisms of GPR. Subsequently, we develop a specific implementation process for our proposed multi-model intelligent prediction method, which takes into account the geological differences at the exploration workface. This method is then applied to actual engineering sites, where we utilize it to forecast and verify the conditions ahead of the tunnel face. The results obtained from these applications are thoroughly discussed to evaluate the method’s effectiveness. Finally, we summarize the key contributions of this research, highlighting its potential for enhancing the safety and efficiency of tunnel construction through more accurate geological predictions.

Comment3. Section 2.1 Rock mass classification is crucial for designing tunnel support. It includes rock mass strength, geological discontinuities conditions, orientation and ground water status. Common classification systems include RMR (Rock Mass Rating) and the Q-system, which use different parameters to classify the rock mass. Authors should use the above mentioned classification of Rock Mass instead of using some report as cited reference 21. 

Reply:

Thank you for your suggestion. We have revised reference [21] to the well - established and widely - used published book, the "Engineering Rock Mass Classification Standard". This standard comprehensively guides rock mass classification for various rock engineering projects. Additionally, we've added detailed introductions to RMR and Q - system along with relevant citations. These adjustments enhance the clarity and credibility of the section.

In 2.1. Rock integrity characteristics and fractured rock structure To quantitatively evaluate rock mass quality and guide tunnel support design, several internationally recognized classification systems have been developed. Among them, the Rock Mass Rating (RMR) system and the Q-system are two widely used approaches [23]. The RMR system evaluates rock mass based on parameters including uniaxial compressive strength, rock quality designation (RQD), joint spacing, joint condition, groundwater inflow, and orientation adjustment [24]. The Q-system, on the other hand, integrates RQD, number of joint sets, joint roughness, joint alteration, stress reduction factor, and groundwater conditions to provide a comprehensive index of rock mass quality [25]. These systems are essential tools for engineers to assess the stability and support requirements of rock masses in various geological settings. Despite the widespread use of RMR and Q-systems, different projects may require specific criteria tailored to local geological conditions. In this study, we adopt the engineering rock mass classification standard as defined by relevant regulations in China [26]. According to this standard, rock integrity is qualitatively classified into five grades: complete, fairly complete, slightly broken, broken, and extremely broken. This classification is based on key indicators such as joint characteristics (including spacing, continuity, and filling materials), the types of dominant structural planes, and their degree of interlocking. These classifications reflect the degree of fracturing and continuity within the rock mass, which directly influence its mechanical behavior and stability. This five-grade classification provides a practical framework for assessing the integrity of rock ahead of tunnel excavation, ensuring appropriate measures are taken to mitigate potential risks during construction.

Comment4. Figure 2 authors mentioned used geological informations in data pre processing, please clarify what are geological informations used?

Reply:

We have clarified in the revised text that the geological information used in data pre - processing is the rock mass integrity grade of the ground - penetrating radar (GPR) operation face. Based on different integrity grades, various sub - models are established to improve the accuracy of the prediction method. This clarification has been added to ensure that readers can clearly understand the data pre - processing process and the basis for sub - model division.

In 3. Methodology ......On this basis, considering the differences in rock mass integrity grades among the GPR exploration working faces, multi-models are established using training data from different degrees of fragmentation on the working face. Finally, according to the working face conditions of the mileage to be predicted, a suitable model will be matched and the rock integrity forecast will be completed.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you to the authors for their revisions and improvements to the article. Personally, I consider it acceptable.

Comments on the Quality of English Language

Verify grammar and expression errors again