A 1/f Noise Detection Method for IGBT Devices Based on PSO-VMD

Round 1

Reviewer 1 Report

The article is about the 1/f noise detection method for IGBT devices based on PSO VMD. From the scientific point of view, the article presents the average scientific level. After the corrections presented for the authors, it is suitable for publication in the journal.

The paper should be improved namely in some identified aspects:

1. There exist some un-defined acronyms through the text, which need to be properly declared.
2. Devices already exist to detect the noise. And it works very well.
3. Section 3.3: Please add the flowchart of the proposed PSO-VMD method? It’s interesting for readers.
4. Lines 67 and 70: the penalty factor is missing.
5. Line 85: correct the number of references. [5] and not ^[5]
6. Concerning the figure 3, please add the legends and correct the sentence: 1/f Noisy 1/f noise signal
7. To conclude the effectiveness of the proposed method, it is worth performing more simulation tests - presenting one result is a weak and controversial way to prove the effectiveness of the proposed method.
8. The novelty of the proposed method should be highlighted to real physics phenomena for fully appreciating its efficiency. This would add value to this work. To be sure that the described method is more effective than other methods, it is necessary to perform an analysis on real signals, measured on real objects.
9. Line 276: why the word detection is in bold?
10. Line 292: please add the references of other algorithms.
11. Line 294: table 3.3 doesn’t exist. Please correct.
12. Moderate English changes are required. Numerous grammatical and spelling errors greatly degraded the quality of the presentation.

 13. The conclusion is missing.  I believe that the "Conclusions" section should be expanded with detailed information about the authors' contribution to solving the issue (authors also use solutions included in other publications), what is new in the presented approach (concerning previously used methods), and what are directions for further research (e.g. measurements on real objects under normal environmental conditions)

Author Response

Response to Reviewer 1 Comments

 

Dear Editors and reviewers:

Thank you for your letter and the reviewers’ comments on our manuscript entitled" A 1/f noise detection method for IGBT devices based on PSO-VMD" (ID：1654618). Those comments are very helpful for revising and improving our paper, as well as the important guiding significance to other research. We have studied the comments carefully and made corrections which we hope meet with approval. The main corrections are in the manuscript and the responds to the reviewers’ comments are as follows (in red).

I am deeply sorry for my late reply. I've been doing volunteer work recently and didn't go home.

 

Point 1: There exist some un-defined acronyms through the text, which need to be properly declared.

 

Response 1: After careful examination, it is found that there are some undefined acronyms in the text. Now the acronyms that appear for the first time have been supplemented except the abstract. These include: BJI, MOSFET,IGBT, EMD, VMD, PSO, IMF, RTN noise and G-R noise, and all have been in red. Thank you.

 

Point 2: Devices already exist to detect the noise. And it works very well.

 

Response 2: IGBT is a high power switch in the field of power electronics. Its reliability is closely related to system stability. Once failure occurs, it may cause irreparable loss.Devices already exist to detect the noise. But it is not suitable for the detection of IGBT low-frequency noise. The reasons are as follows:

1. The low-frequency noise in IGBT devices is usually nV level. Our research group has designed a complete system for IGBT low-frequency noise detection. The system includes detection system, which is divided into hardware and software parts.
2. The hardware system includes low-noise power supply, preamplifier, limited bandwidth filter, low-noise bias circuit and low-noise high-temperature bias circuit. This part is responsible for the detection of drain voltage noise in IGBT devices.
3. The software part analyzes the data obtained, including the time series and spectrum analysis of low-frequency noise, intuitively displays the time series curve of noise, then analyzes the spectrum of low-frequency noise according to the data, and finally realizes the detection of low-frequency noise of IGBT devices.

 

Point 3: Section 3.3: Please add the flowchart of the proposed PSO-VMD method? It’s interesting for readers.

 

Response 3: This is a good advice. We have added Figure 2. PSO-VMD detection 1/f noise flow chart.

Point 4: Lines 67 and 70: the penalty factor is missing.

 

Response 4: Penalty factors have been added.

 

Point 5: Line 85: correct the number of references. [5] and not [5]

 

Response 5: It has been corrected.

 

Point 6: Concerning the figure 3, please add the legends and correct the sentence: 1/f Noisy 1/f noise signal

 

Response 6: This has been corrected to Figure 4. 1/f noise signal.

 

Point 7: To conclude the effectiveness of the proposed method, it is worth performing more simulation tests - presenting one result is a weak and controversial way to prove the effectiveness of the proposed method.

 

Response 7: Lines 85 and 86: For IGBT modules, material failure or imperfect manufacturing process will cause 1 /f noise, which is a main low-frequency noise reflecting the performance of IGBT.

Lines 93 to 95: Therefore some researches showed that structural defects of BJT and MOSFET, such as microtubules, dislocations, doping, carrier lifetime and stacking, will affect their reliability, and defects will cause 1/f noise [22].

 

Point 8: The novelty of the proposed method should be highlighted to real physics phenomena for fully appreciating its efficiency. This would add value to this work. To be sure that the described method is more effective than other methods, it is necessary to perform an analysis on real signals, measured on real objects.

 

Response 8: Lines 216 to 266，we have added relevant research contents.

Figure3. IGBT low frequency noise detection scheme

Figure4. IGBT low frequency noise test environment

 

Point 9: Line 276: why the word detection is in bold?

 

Response 9: Bold has been canceled for  detection .

 

Point 10: Line 292: please add the references of other algorithms.

 

Response 10: Lines 355 to 357: Improved EMD algorithm, Heuristic threshold, Minimax threshold, Lifting wavelet threshold, Wavelet entropy soft threshold, Lifting wavelet decomposition wavelet entropy soft threshold.

 

 

Point 11: Line 294: table 3.3 doesn’t exist. Please correct.

 

Response 11: Corrected to Table 4.

 

Point 12: Moderate English changes are required. Numerous grammatical and spelling errors greatly degraded the quality of the presentation.

 

Response 12: According to your opinion, we have made a lot of changes.

 

Point 13: The conclusion is missing. I believe that the "Conclusions" section should be expanded with detailed information about the authors' contribution to solving the issue (authors also use solutions included in other publications), what is new in the presented approach (concerning previously used methods), and what are directions for further research (e.g. measurements on real objects under normal environmental conditions)

 

Response 13: We added a photo of the experimental environment. Fig. 4 shows the experimental hardware platform established based on IGBT low frequency noise test environment.

Lines 377 to 381: The signal power spectral density has obvious characteristics of 1 / f noise signal, and its power spectral density decreases with the increase of frequency. When using pso-vmd algorithm to process 1 / f noise signal, the extracted signal has the same smoothness as the original signal, and its power spectrum is more accurate.

 

Once again, thank you very much for your constructive comments and suggestions which would help us both in English and in depth to improve the quality of the paper.

 

Kind regards,

 

Jie Wu

 

E-mail: bhwj@beihua.edu.cn

 

Corresponding author: Xiaojuan Chen

 

E-mail address: 2016200069@mails.cust.edu.cn

 

Reviewer 2 Report

A review of the paper can be found in the attached file.

Comments for author File: Comments.pdf

Author Response

Response to Reviewer 2 Comments

 

Dear Editors and reviewers:

Thank you for your letter and the reviewers’ comments on our manuscript entitled" A 1/f noise detection method for IGBT devices based on PSO-VMD" (ID：1654618). Those comments are very helpful for revising and improving our paper, as well as the important guiding significance to other research. We have studied the comments carefully and made corrections which we hope meet with approval. The main corrections are in the manuscript and the responds to the reviewers’ comments are as follows.

I am deeply sorry for my late reply. I've been doing volunteer work recently and didn't go home.

 

Remarks: Very many stylistic errors were noted.

We carefully checked and revised a lot stylistic errors.

 

Suggestions:

1) Please complete explanations of all symbols and shortcuts that appear in the text.

After careful examination, it is found that there are some undefined acronyms in the text. Now the acronyms that appear for the first time have been supplemented except the abstract. These include: BJI, MOSFET, EMD, VMD, PSO, IMF, RTN noise and G-R noise, and all have been in red. Thank you.

 

2) I suggest aligning both equation numbering and symbols evenly with the text.

We have revised equation numbering and symbols carefully.

 

Once again, thank you very much for your constructive comments and suggestions which would help us both in English and in depth to improve the quality of the paper.

 

Kind regards,

 

Jie Wu

 

E-mail: bhwj@beihua.edu.cn

 

Corresponding author: Xiaojuan Chen

 

E-mail address: 2016200069@mails.cust.edu.cn

 

Author Response File: Author Response.doc

Reviewer 3 Report

This paper investigated in PSO-VMD algorithm to extract the 1/f noise from white noise. By simulating under different signal-to-noise ratios, the authors show that the proposed PSO-VMD algorithm realized the most effective separation of the components of the detection signal compared with the other 6 algorithms. This manuscript can be accepted for publication after minor revision.

1. Table 3 and table 4 are the same. As well, “Table III” is mentioned in line 278, “Table 3-3” in line 285 and 294. This is confusing. Please try to them clear.
2. In lines 214, 217 and 230, the authors are using “Fig 2” “Fig 4” or “Figure 3”. Please keep them consistent in the whole manuscript.

Author Response

Response to Reviewer 3 Comments

 

Dear Editors and reviewers:

Thank you for your letter and the reviewers’ comments on our manuscript entitled" A 1/f noise detection method for IGBT devices based on PSO-VMD" (ID：1654618). Those comments are very helpful for revising and improving our paper, as well as the important guiding significance to other research. We have studied the comments carefully and made corrections which we hope meet with approval. The main corrections are in the manuscript and the responds to the reviewers’ comments are as follows.

I am deeply sorry for my late reply. I've been doing volunteer work recently and didn't go home.

 

1. Table 3 and table 4 are the same. As well, “Table III” is mentioned in line 278, “Table 3-3” in line 285 and 294. This is confusing. Please try to them clear.

We have made corresponding changes, and thank you for your kind proposal! 

 

2. In lines 214, 217 and 230, the authors are using “Fig 2” “Fig 4” or “Figure 3”. Please keep them consistent in the whole manuscript.

We have made corresponding changes, and thank you for your kind proposal!

 

Once again, thank you very much for your constructive comments and suggestions which would help us both in English and in depth to improve the quality of the paper.

 

Kind regards,

 

Jie Wu

 

E-mail: bhwj@beihua.edu.cn

 

Corresponding author: Xiaojuan Chen

 

E-mail address: 2016200069@mails.cust.edu.cn

 

Author Response File: Author Response.doc

Reviewer 4 Report

This paper focuses on the detection method of 1/f noise signal based on VMD using PSO algorithm. The method presents the enhanced detection capability. The detailed questions are provided below.

1. The abstract is recommended to include full names of IGBT, IMF, VMD, and PSO.

2. The originality should be clarified, whether this paper is the first case applying PSO algorithm to VMD model. If not, a noticeable difference from previous works should be mentioned.

3. The reason why the authors choose IGBT devices to apply PSO-VMD model is recommended to be mentioned, while there are many electronic devices suffering from 1/f noise.

Author Response

Response to Reviewer 4 Comments

 

Dear Editors and reviewers:

Thank you for your letter and the reviewers’ comments on our manuscript entitled" A 1/f noise detection method for IGBT devices based on PSO-VMD" (ID：1654618). Those comments are very helpful for revising and improving our paper, as well as the important guiding significance to other research. We have studied the comments carefully and made corrections which we hope meet with approval. The main corrections are in the manuscript and the responds to the reviewers’ comments are as follows (in red).

I am deeply sorry for my late reply. I've been doing volunteer work recently and didn't go home.

 

1. The abstract is recommended to include full names of IGBT, IMF, VMD, and PSO.

After careful examination, it is found that there are some undefined acronyms in the text. Now the acronyms that appear for the first time have been supplemented except the abstract. These include: BJI, MOSFET, EMD, VMD, PSO, IMF, RTN noise and G-R noise, and all have been in red. Thank you.

2. The originality should be clarified, whether this paper is the first case applying PSO algorithm to VMD model. If not, a noticeable difference from previous works should be mentioned.

Our paper isn’t the first case applying PSO algorithm to VMD model. However, it is innovative to use this method to process low-frequency noise signals. We have added the corresponding reference.

3. The reason why the authors choose IGBT devices to apply PSO-VMD model is recommended to be mentioned, while there are many electronic devices suffering from 1/f noise.

IGBT is a high power switch in the field of power electronics. Its reliability is closely related to system stability. Once failure occurs, it may cause irreparable loss.Devices already exist to detect the noise. But it is not suitable for the detection of IGBT low-frequency noise. The reasons are as follows:

The low-frequency noise in IGBT devices is usually nV level. Our research group has designed a complete system for IGBT low-frequency noise detection. The system includes detection system, which is divided into hardware and software parts.

The hardware system includes low-noise power supply, preamplifier, limited bandwidth filter, low-noise bias circuit and low-noise high-temperature bias circuit. This part is responsible for the detection of drain voltage noise in IGBT devices.

The software part analyzes the data obtained, including the time series and spectrum analysis of low-frequency noise, intuitively displays the time series curve of noise, then analyzes the spectrum of low-frequency noise according to the data, and finally realizes the detection of low-frequency noise of IGBT devices.

 

Once again, thank you very much for your constructive comments and suggestions which would help us both in English and in depth to improve the quality of the paper.

 

Kind regards,

 

Jie Wu

 

E-mail: bhwj@beihua.edu.cn

 

Corresponding author: Xiaojuan Chen

 

E-mail address: 2016200069@mails.cust.edu.cn

 

Author Response File: Author Response.doc

Round 2

Reviewer 1 Report

The comments are dealt with satisfactory

Reviewer 4 Report

I recommend that this paper be accepted.