Prediction and Optimization of Blasting-Induced Ground Vibration in Open-Pit Mines Using Intelligent Algorithms

Round 1

Reviewer 1 Report

In this work, a hybrid intelligent model based on a least-squares support vector machine and particle swarm algorithm to predict the ground vibrations. The blast design was optimized be using multi-objective particle swarm optimization. The proposed agathism has been verified and applied to optimize the blast design. Overall, the work is new, original, high applicability and valuable for publication in Applied Sciences. Besides, the following major issues should be addressed:

 

1) The novelty of the current study should be more discussed in the Abstract section.

2) The manuscript should be checked and re-presented to improve the quality of the presentation. The following suggestions should be considered:

- The current form of table 2 is not good, it should be revised (Some numbers are broken into 2 lines).

- Table 3 should be changed to fit the format of the journal and made it easier to read.

- The size of Fig. 10 and Fig. 11 should be the same.

- The matrix and vector should change to bold front (for example Eq. (7) and its explanations and other positions).

3) The following up-to-date references should be considered and discussed to enhance the brief-review of published data:

- https://doi.org/10.3390/app10041403

- https://doi.org/10.1007/s11053-018-9424-1

4) Kernel selection: LS-SVM employs kernel functions to transform the input data into a higher-dimensional feature space. It is important to choose an appropriate kernel based on the problem at hand. Common choices include linear, polynomial, radial basis function (RBF), and sigmoid kernels. The choice of the kernel can significantly impact the model’s performance. The author is suggested to more explain about the chosen of radial basis function (RBF) in Eq. (10)?

5) Conclusion: The conclusion section should be revised. The application of the present method and important results should be discussed in conclusion section. Some suggestions for further studies should be given in conclusion.

The English language is good.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

The authors used PPV as a vibration index. Why did not the authors use RMS of velocity? PPV corresponds to a maximum value of vibrations (it occurred instantly). However, RMS corresponds to the average of time history of vibrations. 

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

This paper presents the peak particle velocity (PPV) prediction in blasting with PSO-LSSVM algorithm for less prediction error of root-mean-square-error 19 (RMSE) and the mean absolute error (MAE), and better correlation coefficient (r), compared with genetic algorithm optimized BP neural network (GA-BP), unoptimized LSSVM, and BP.  Furthermore, the blasting parameters were optimized through  Multi-objective PSO (MOPSO). It is good that this paper studies the side effect of blasting mines through advanced PSO-LSSVM and MOPSO algorithms for better blasting effect prediction and blasting parameters optimization. 

However, the contribution of this paper is not clearly stated in the introduction and some information is missed in the methods.

For overall recommendation, this paper will be good to be published with some minor revisions.

My detailed questions/comments/suggestions on the content:

1. What is the contribution of this paper? What is the innovation in this paper? What makes this paper different from previous works? These questions should be answered in the introduction. Although the author reviewed a lot of previous work in the introduction, the contrition and impact of this paper is not mentioned in the introduction.
2. Could you mark the blasting parameters like monitoring distance (R), negative elevation (H), Burden (B), spacing (S), Subdrilling (H0), etc. clearly in Fig 2 and Fig.3 so that the readers will know exactly where they are in the experiment?
3. Did you verify the optimized parameters with the experiment? 

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

All issues have been addressed. The manuscript can be accepted in the current form.