Application of an Adaptive “Neuro-Fuzzy” Inference System in Modeling Cutting Temperature during Hard Turning

Round 1

Reviewer 1 Report

The paper presents the application of adaptive neuro-fuzzy inference system in machining process. The problems of an optimization of hard cutting processes have been previously extensively investigated by many researchers, however the advantage of this work is the development of an interesting adaptive neuro-fuzzy inference system. In a current form, the paper contains some flaws, thus in order to accept, some major revision is required. The detailed remarks are as follows:

The paper is focused on the evaluation of adaptive neuro-fuzzy inference system in relation to cutting temperatures generated during hard turning. Therefore, the current version of manuscript’s title is not fully reflecting the content of paper. Thus, the title should be modified. In abstract: the Authors claim that: “Machining of hardened steel by turning is a relatively new process that uses advanced machines and tools”. Currently, this statement is untrue, since the hard machining technology is developing since at least two decades! The state-of-the-art presented in paper is well organized. However, some latest works regarding the mechanical and technological aspects of hard machining have been omitted. Therefore, the Introduction section should be extended and include the discussion of following works: Surface quality and topographic inspection of variable compliance part after precise turning. Applied Surface Science. Volume 434, 15 March 2018, Pages 91-101. Intelligent optimization of hard-turning parameters using evolutionary algorithms for smart manufacturing. Materials 16 (6),879, 2019. Mechanical and technological aspects of micro ball end milling with various tool inclinations. International Journal of Mechanical Sciences. Volume 134,  2017, Pages 424-435. The cutting temperature is defined as a temperature generated in a cutting zone. According to Authors, they applied the thermal camera, which is unable the measure the direct cutting temperature. Therefore, please provide more details, regarding this measurement method, i.e. in which location in tool-workpiece system, the measurement has been made? It is also advised to add the appropriate figure with the detailed scheme of measurement. The conducted analysis of results is not fully satisfying. It concentrates mainly on the evaluation of ANIFS model instead of relations between the process’ input parameters, types of tool and the obtained values of temperature. Since the paper is focused on machining of hardened steels in terms of temperature, it is very important to conduct the in-depth analysis of relations between the input parameters and temperature, i.e. please explain why the values of cutting temperature are lower during cutting with CBN than with carbide insert? Why the depth of cut has non-monotonic influence on the generated temperature?

Author Response

Review 1

The paper is focused on the evaluation of adaptive neuro-fuzzy inference system in relation to cutting temperatures generated during hard turning. Therefore, the current version of manuscript’s title is not fully reflecting the content of paper. Thus, the title should be modified

Response:

The title has been changed.

Application of adaptive neuro-fuzzy inference system in modeling cutting temperature during hard turning

In abstract: the Authors claim that: “Machining of hardened steel by turning is a relatively new process that uses advanced machines and tools”. Currently, this statement is untrue, since the hard machining technology is developing since at least two decades.

Response: The sentence was replaced.

Increasing demands of the market require a hard processing hardened steel in order to avoid finishing grinding.

The state-of-the-art presented in paper is well organized. However, some latest works regarding the mechanical and technological aspects of hard machining have been omitted. Therefore, the Introduction section should be extended and include the discussion of following works: Surface quality and topographic inspection of variable compliance part after precise turning. Applied Surface Science. Volume 434, 15 March 2018, Pages 91-101. Intelligent optimization of hard-turning parameters using evolutionary algorithms for smart manufacturing. Materials 16 (6),879, 2019. Mechanical and technological aspects of micro ball end milling with various tool inclinations. International Journal of Mechanical Sciences. Volume 134,  2017, Pages 424-435.

Response:

The literature suggested by the reviewer has been introduced.

 

The cutting temperature is defined as a temperature generated in a cutting zone. According to Authors, they applied the thermal camera, which is unable the measure the direct cutting temperature. Therefore, please provide more details, regarding this measurement method, i.e. in which location in tool-workpiece system, the measurement has been made? It is also advised to add the appropriate figure with the detailed scheme of measurement.

Response:

Figure 1. Experiment setup a) A view of the thermal camera when measuring b) A thermal imaging with measurement position

It concentrates mainly on the evaluation of ANIFS model instead of relations between the process’ input parameters, types of tool and the obtained values of temperature.

Response:

It is evident from figure 4 that the increase in cutting speed is reflected by an increase in the cutting temperature. The reason for the increase in cutting temperature is due to the conversion of mechanical energy (rotation of spindle) into heat energy. Cutting speed with the CBN tool has a similar effect, figure 5. The effects of feed on the cutting temperature for both tool material were contrary to expectations. ….

Compared to literature sources, the cutting temperatures obtained in this paper are relatively lower. This happens by adopting a medium cutting speed according to the manufacturer's recommendations and a way to measure temperature. The use of thermocouples gives a more realistic temperature, but on the economic side it is quite expensive.

Figure 4. Main effect plot for cutting temperature with HM tool

Figure 5. Main effect plot for cutting temperature with CBN tool

Since the paper is focused on machining of hardened steels in terms of temperature, it is very important to conduct the in-depth analysis of relations between the input parameters and temperature, i.e. please explain why the values of cutting temperature are lower during cutting with CBN than with carbide insert ? Why the depth of cut has non-monotonic influence on the generated temperature?

Response:

For instance, an increase in feed rate caused a significantly increase in cutting temperature. Lastly, depth of cut also has an effect on the cutting temperature, especially when turning with a HM tool, while with CBN it showed a smaller effect. This can be explained by the fact that the tool material of cubic boron nitride has better thermal conductivity. This was also confirmed by experiments that showed that generally higher temperatures occur with HM tools. The maximum temperature at HM was 350 ̊C while at CBN was 169 ̊C.

Author Response File: Author Response.doc

Reviewer 2 Report

Please provide the references for the 1^st sentence of 1^st paragraph in the section of Introduction and indicate the aims in the current study. The authors should rewrite this section. In addition, the current literature review is not sufficient, please mention the more studies in the recent years, such as 2019, 2018, 2017.

 

Cubic boron nitride (CBN) and hard metal (HM) have been mentioned in the abstract section, so please just use CBN and HM in the text, like line 135.

 

Could you please provide more details about the experiment?

 

 

Could the authors compare the current results and the previous studies? How to proof the current method is better than the previous studies?

Author Response

Review 2

Please provide the references for the 1^st sentence of 1^st paragraph in the section of Introduction and indicate the aims in the current study.

Response: The reference was added to the first sentence.

The authors should rewrite this section. In addition, the current literature review is not sufficient, please mention the more studies in the recent years, such as 2019, 2018, 2017.

Response: Newer references are inserted in the introduction.

Cubic boron nitride (CBN) and hard metal (HM) have been mentioned in the abstract section, so please just use CBN and HM in the text, like line 135.

Response: Throughout the text, abbreviations HM and CBN were used.

Could you please provide more details about the experiment?

Response: The figure 1 provide more details about experiment.

Could the authors compare the current results and the previous studies? How to proof the current method is better than the previous studies?

Response: The introduction briefly outlines the benefits of the ANFIS technique over other AI tools.

Author Response File: Author Response.doc

Reviewer 3 Report

Acceptable with minor spell checking.

Author Response

Review 3

English language and style are fine/minor spell check required.

Response: Errors found throughout the text have been corrected.

Author Response File: Author Response.doc

Reviewer 4 Report

An interesting article on the use of advanced neural networks to analyze the results of experimental research. I have some advices for authors of this article

Comment 1
The title of article is too extensive research on the subject of the proposed title I proposel to limit the title to the scope of research  for example.

"Application of adaptive neuro-fuzzy inference system 2 in turning process for prediction of cutting temperature"

Comment 2

The review of the literature should be extended to include examples of research into the cutting process of recent years, for example on the recognition of the wear of the edge on the basis of the image of wear
( Neural network approach for automatic image analysis of cutting edge wear). or the prediction of the tool life (ex.  Predicting tool life in turning operations using neural networks and image processing). In paper was described only examples of ANFIS. Comparison of this solution can show of benefits ANFIS

Comment 3

Fig. 4 shows a graphic comparison of dependencies because it was implemented for other scales, it does not reflect actual changes. It proposes to present the results in the same vertical scale.

Comment 4

If author agree it would be interesting to compare the relative changes compared to the central point of the experiment

Comment 5

The obtained temperature values ​​seem to be quite low whether it results from the adopted measurement technique or too low cutting speeds ?.

Comment 6

The authors do not explain why the depth of cut is different for CBN than HM (decreases with increase of ap)

 

After review an article it possible it publication

Author Response

Review 4

The title of article is too extensive research on the subject of the proposed title I proposel to limit the title to the scope of research for example. "Application of adaptive neuro-fuzzy inference system 2 in turning process for prediction of cutting temperature"

Response: The title has been changed.

 

The review of the literature should be extended to include examples of research into the cutting process of recent years, for example on the recognition of the wear of the edge on the basis of the image of wear ( Neural network approach for automatic image analysis of cutting edge wear). or the prediction of the tool life (ex. Predicting tool life in turning operations using neural networks and image processing). In paper was described only examples of ANFIS. Comparison of this solution can show of benefits ANFIS

Response:

In recent years, a whole range of Artificial Intelligence (AI)-based techniques have been developed that model the correlation between the input (process data, like cutting speed, feed and depth of cut) and the output (tool life, cutting temperature, surface quality, etc.) parameters of the turning process. Mia et al. have dealt with optimization of hard-turning parameters using evolutionary algorithms [19]. In their work investigated machining parameters were cutting speed, feed rate, and depth-of-cut, while the output performance were surface roughness and cutting temperature. An example of a cutting temperature prediction for the turning of biomedical stainless steel is carried out by Petkovic et al. [20]. These authors used a neural network with an optimization algorithm for quick, easy and successful optimization of input regimes. Consequently, the intelligent model on the basis of the neural network is a good feature of generalization and has the ability to accept nonlinear variables with unknown iterations. Mikołajczyk et al. predicted tool life tool life in turning operations using neural networks [21]. Their results confirm that the combination image recognition software and ANN modeling could potentially grow into a useful industrial tool for cheap estimation of tool life in agile operations. One example of the application of fuzzy logic is shown by Prabhu et al. [22]. They used fuzzy logic analysis method to predict the best optimal solution, and to find out the most influential parameter to determine the output characteristics.

 

4 shows a graphic comparison of dependencies because it was implemented for other scales, it does not reflect actual changes. It proposes to present the results in the same vertical scale.

Response: The figure is drawn at the same scaling of temperature.

If author agree it would be interesting to compare the relative changes compared to the central point of the experiment

Response:

The hard-turning operation with both tool materials are controlled by three factors, like the cutting speed, feed rate, and depth of cut. Therefore, these parameters generally affect the possible results of the machining. For that reason, the mean behaviors of the cutting temperature are evaluated by dispersion analysis and main effect plot. The main effect plot show a relative change from the center point of the experiment.

The obtained temperature values seem to be quite low whether it results from the adopted measurement technique or too low cutting speeds

Response:

Compared to literature sources, the cutting temperatures obtained in this paper are relatively lower. This happens by adopting a medium cutting speed according to the manufacturer's recommendations and a way to measure temperature. The use of thermocouples gives a more realistic temperature, but on the economic side it is quite expensive.

The authors do not explain why the depth of cut is different for CBN than HM (decreases with increase of ap)

Response:

This can be explained by the fact that the tool material of cubic boron nitride has better thermal conductivity. This was also confirmed by experiments that showed that generally higher temperatures occur with HM tools. The maximum temperature at HM was 350 ̊C while at CBN was 169 ̊C.

Author Response File: Author Response.doc

Round 2

Reviewer 1 Report

The Authors, have carefully considered the reviewers' remarks in the revised paper. In my opinion, its quality has now been significantly improved and thus it can be accepted for publication.

Author Response

Review 1: English language and style are fine/minor spell check required

Response: Spell check English and style was done.

Author Response File: Author Response.doc

Reviewer 2 Report

4 and 5 can be combined to one figure with sub-caption. In addition, please provide sub-caption for sub-figure in Figs. 6 and 7.

Author Response

Review 2: Figure 4 and 5 can be combined to one figure with sub-caption. In addition, please provide sub-caption for sub-figure in Figs. 6 and 7.

Response: All pictures are arranged according to the instructions of the author.

Author Response File: Author Response.doc