Fault Diagnosis for Body-in-White Welding Robot Based on Multi-Layer Belief Rule Base

Round 1

Reviewer 1 Report

Report: Fault Diagnosis for Body-In-White Welding Robot Based on Multi-Layer Belief Rule Base

Summary:

The paper presents a fault diagnosis model for body-in-white (BIW) welding robots based on a multi-layer belief rule base (BRB). The authors propose constructing the model using fault tree analysis, incorporating both monitoring data and expert knowledge. They use the projection covariance matrix adaptive evolutionary strategy (P-CMA-ES) algorithm to optimize the model parameters. The proposed model is validated through a simulation experiment on a BIW welding robot, demonstrating improved diagnostic accuracy compared to other models.

Major Revision Suggestions:

Context and motivation:

The article do not adequately emphasize the significance of the problem and the novelty of the proposed approach. The introduction should provide more context on the importance of fault diagnosis in BIW welding robots, discuss the current state of the art, and highlight the limitations of existing methods that the proposed model seeks to address.

Methodology details:

The article provide only a high-level overview of the methodology. The paper should include a detailed description of the fault tree construction, the multi-layer BRB model development, and the P-CMA-ES algorithm. The authors should explain the rationale behind their choices and how each component contributes to the model's performance.

Comparison with existing methods:

The conclusion mentions a comparison with other models but does not provide any specific details. The paper should include a comprehensive comparison of the proposed method with existing fault diagnosis techniques, supported by quantitative metrics and clear criteria for evaluation. This comparison will help demonstrate the advantages of the proposed model and justify its adoption.

Validation and experiments:

The abstract mentions simulation experiments to validate the proposed model, but the conclusion lacks details on these experiments. The paper should elaborate on the experimental setup, the data used, the performance metrics employed, and the specific results obtained. This information will help readers assess the effectiveness and robustness of the proposed method.

Discussion and limitations:

The paper should include a discussion section that analyzes the results, explores the implications of the findings, and identifies potential limitations of the proposed method. This section should also suggest possible improvements or future research directions to address these limitations.

Clarity and organization:

The paper would benefit from a clearer structure and more precise language. The authors should ensure that each section is well-organized, with a logical flow of ideas, and that the terminology and concepts are explained clearly and concisely. Additionally, the paper should be thoroughly proofread to eliminate any grammatical errors or inconsistencies. The plagiarism of the paper is 28% it should be less than 20 or at least 25 %.

Author Response

The modification of the article has been uploaded in the Word file below

Author Response File: Author Response.pdf

Reviewer 2 Report

The article presents a methodological proposal for the fault diagnosis of the body-in-white welding robot, under the conditions of uncertainty and the complexity of achieving this task, a method based on multi-layer belief rule base and the projection covariance matrix adaptive evolutionary strategy, finally the authors present simulation results. This article is of current interest since one of the main problems in fault diagnosis is the uncertainty in the parameters. here some comments.

1.      It is recommended to emphasize the main contributions of the paper in the introduction of the article, in order to clarify its content.

2.      it is recommended to increase the quality of figure 2, since it looks blurry

3.      generally the initial parameters of a BRB model are based on empirical knowledge by the expert, which are the main reasons why the authors preferred to use Projection Covariance Matrix Adaptation Evolution Strategy?

4.      In the data shown in Graph 6, what is the probability distribution to the random variables shown?

5.      minor revision in spelling to the text is recommended.

6.      It is suggested that in the conclusions the possible future extensions of the work presented be mentioned, as well as the gabs filled by the proposal in reference to other related works in the area.

Author Response

Please refer to the Word file uploaded below

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors revised the manuscript.