Hybrid Modeling and Simulation of the Grinding and Classification Process Driven by Multi-Source Compensation

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In this paper, the hybrid modeling and simulation of grinding classification process through Multi-source Compensation is of great significance, which can provide good theoretical and practical value for intelligent control of grinding classification process In my opinion the paper may be reconsidered for publication after minor modifications, following are my specific comments:

Point 1: The abstract should be re-refined to supplement the research background and the application results of the research findings.

Point 2: 5 Conclusions should be changed to modify the model for industrial applications.

Point 3: The concentrator situation and industrial process for which the model is applied should be clearly described in 5.1.

Point 4: The fitting results of the parameters related to the feed and pump mentioned above are recommended in Section 5.2.

Comments on the Quality of English Language

No

Author Response

Response to Reviewer 1 Comments

We feel great thanks for your professional review work on our paper. As you are concerned, there are several problems that need to be addressed. According to your nice suggestions, we have made extensive correction to our previous draft. We look forward to hearing from you regarding our submission. We would be glad to response to any further questions and comments that you may have.

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

Reviewer: The article effectively addresses the problem of modeling the ore grinding and classification system using a heuristic approach and artificial intelligence. The authors understand the specifics of the research object, its limitations and the dynamic nature of the process. The formulation of the research problem and the methodology for its solution are correct. The article can be accepted for publication after taking into account the comments below.

Specific comments:

1. Please explain the letter designations in formula 7

2. Please explain the letter designations y1,2,3,4, and T in the Y compensation model.

3. Please expand the abbreviation of the DBDAE algorithm.

4. Line 154: The obtained hidden layer is the input nonlinear feature vector. Shouldn't it be …output…

5. Please correct the quality of Figure 4

6. Chapters 5 and 6 have the same title

7. In the text, the authors inform about the measured process parameter: of particle content under 74 micron. Above the graph (Fig. 5) is the caption -200?

8. In Figures 5-7 is the graph in black, described in the text as red?

9. Please explain why with the decrease in the flow of ore to the mill (t/h), the percentage of -74 micron particles in the hydrocyclone overflow (Fig. 9-10) decreased? The trend should be different. A less loaded mill will grind more efficiently, there should be more % of -74 micron particles, and more % of +74 micron particles.

10. Please provide the quality coefficients of training, testing and validation of the neural network. R2 coefficient as the degree of explanation of the model by the process variables.

Author Response

We feel great thanks for your professional review work on our paper. As you are concerned, there are several problems that need to be addressed. According to your nice suggestions, we have made extensive correction to our previous draft. We look forward to hearing from you regarding our submission. We would be glad to response to any further questions and comments that you may have.

Please see the attachment.

Author Response File: Author Response.docx