The Design of Reflected Laser Intensity Testing System and Application of Quality Inspection for Laser Cladding Process
Round 1
Reviewer 1 Report
This paper proposes a muti-signal test system for measuring the process variables while laser cladding, and uses machine learning to predict the cladding quality. The paper presents a preliminary results by researching limited experimental data. The manuscript may be major revised before acceptance. The comments are as follows:
1. The paragraph in Line 70-77 may be revised as “in this paper, …”.
2. The meaning of the paragraph in line 93-96 is ambiguous, and the paragraph may be divided into more paragraphs with a smaller length.
3. Why is it possible to evaluate or feedback on the partial cladding quality through the three groups of data?
4. In the test experiment, how to receive the reflected laser in the particular direction. In my point of view, there may be several reflected laser in different directions.
5. What’s the meaning of the sentence “…irradiates the metal surface for 4ms every 0.1s to form a cladding spot.” Also, duty ratio is rarely used to describe the laser pulse parameters, duty ratio may be replaced by the pulse duration.
6. The details of processing data in section 3.3 may be describe in depth.
7. Were the laser cladding parameters changed in the experiments, or only one set of parameters were employed to fabricate the test sample.
Comments for author File: 
Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
The design of the control system or the interpretation of the measurement signals is based on equation 2. The integral over the temperature with respect to the heat capacity, which is supposed to represent the absorbed energy, makes no sense. The volume integral over the component is still missing! The integral over the solid angle results in the unit J/s=W and the integral over the temperature J. So this does not fit. If you want to approximate the integral over the solid angle over 6 interpolation points, this is not representative of the complete integral in my view. The interaction of the laser radiation with the powder is not addressed here at all?! During pulsed processing, powder particles can accumulate in front of the fusion line and lead to ejections or defects in the bond during the next pulse in overlap processing, as can be observed in high-speed images, for example. The problem or the process complexity is not dealt with adequately in this article. The complexity of the test set-up is not even close to practical relevance if high-quality component repair welding is to be carried out. The layers are usually applied to free-form surfaces.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
The author´s manuscript was based on solid experimental results. They proposed and built a test system for measuring the process variables for laser cladding experiments. The manuscript provided also theoretical analysis on transfer conversion of laser energy. Their laser cladding test results were discussed in detail. This manuscript will be very useful for laser cladding researchers and may attracts many reads and most hopefully citations for “Machines”. I therefore would like to recommend the acceptance of this manuscript for possible publication by “Machines”
Not very original in relation to previous published articles, however, the authors have built experimental prototype, several important parameters were investigated by machine learning to study the correlation between process variables and cladding quality, the measured data and cladding results are discretized and quantified, which demonstrates the author´s originality.
Compared with other published material，Machine Learning to study the correlation between process variables and cladding quality, the measured data and cladding results are discretized and quantified.
Specific improvements:By analyzing more laser welding parameters, their methods can be more convincing.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
This paper could be accepted in the present form.
