Tool Wear Prediction Based on Residual Connection and Temporal Networks

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1. In section 2.1.1, the diagram of the structure of CNN should be given.

2. The original 7 channels of signals, the corresponding extracted features, and the corresponding tool wear, should be given for different degree of tool wear, for illustrating the experiment more clearly.

3. The PHM2010 dataset is used for model verification, and the experiment condition is described, the result show that the proposed model is effective. But the paper needs to supplement more experiment to verify the effectiveness of the proposed model, especially the actual cutting experiment.

Comments on the Quality of English Language

no suggestion

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

1. Can you explain the concept of non-uniform material (line 14)

2. Justify the choice of cutting conditions

3. Describe in detail the technological system - machine-tool-preparation-material

4. Describe the milling methodology in the experiment

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Dear authors, the work has interesting results and can contribute to the field, however, there are some important points that need to be addressed prior to acceptance.

 

Abstract: avoid colloquial language like “cutting mileage”, also is strange to say that the tool wear degrades, the tool wear increase, the quality of part decrease, etc.

All the abstract is too confuse, needs to be re-written.

 

Introduction

The contribution of each reference must be cited independently, avoid too long paragraphs.

Improve figure 4, the others a better presented and more readable.

 

The experimental procedure should be carefully revised.

there are some translation issues, it should be used rpm, radial and longitudinal depth of cut, not proper to use the tool rotation and leaving to the reader to estimate the cutting velocity.

The cutting parameters are note clear, was that only one tool? Cutting fluid applied? Cutting length between the experiments?

there are no data about the sensors used, only the manufacturer, it is not proper, all the details must be provided, the ranges of the equipment, if there were amplifiers, filters used… were the machining frequencies on the sensors ranges?

Does the equipment selected possess resolution to obtain the machining variations? The signals are the main input of the work, and the acquisition should be properly determined and described.

 

On item 4.2, does channel means the acquisition channel or a slot manufactured? If so, why 7 and not 6 as described?

Which data processing was used? There are filters that alters the frequency response, more details should be added.

Is an “acceleration sensors” different from an accelerometer?

 

Results and discussion should be item 5

There are no discussions, the results are interesting, but there is the need to be better described and there is no discussion, it is not proper to only present 1 comparison to other works, it should be added.

The conclusions should be revised after the corrections are made.

Comments on the Quality of English Language

Mentioned on the revision.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf