Open-Circuit Fault Diagnosis of Wind Power Converter Using Variational Mode Decomposition, Trend Feature Analysis and Deep Belief Network

Round 1

Reviewer 1 Report

I strongly suggest you to completely revise the paper from the English point of view. Moreover, bibliography and experimental results should be improved.

The submitted article mainly focuses on the proposal of a rather original (NNs-based) method for the diagnosis of both single and double insulated gate bipolar transistor (faults) with respect to the functioning of (specific elements) wind turbines. Without entering into details, the proposed study is very applied, not sure about its interest within a wider community other than the one strictly related to wind turbine functioning. On of the major lacks of the article concerns its English version. Surely, it has to be rewritten, possibly with the help of a English native speaker.

Proposed method should be compared to other (even classical ones) to better highlight the claimed high level in accuracy.

I would suggest (also in view of the previous remark) to include at least the following references:

• Lu, B., Sharma, S.K. A literature review of IGBT fault diagnostic and protection methods for power inverters (2009) IEEE Transactions on Industry Applications, 45 (5), pp. 1770-1777.
• Di Persio, L., Honchar, O. Analysis of recurrent neural networks for short-term energy load forecasting (2017) AIP Conference Proceedings, 1906, art. no. 190006, .

* Wang, H., Zhang, C., Zhang, N., Chen, Y., Chen, Y. Fault Diagnosis for IGBTs Open-Circuit Faults in High-Speed Trains Based on Convolutional Neural Network (2019) 2019 Prognostics and System Health Management Conference, PHAI-Qingdao 2019, art. no. 8943008, .

• Kuraku, N.V.P., He, Y., Ali, M. Fault diagnosis of open circuit multiple IGBT’s using PPCA-SVM in single phase five-level voltage controlled H-bridge MLI (2020) IEEJ Journal of Industry Applications, 9 (1), pp. 61-72.
• Li, B., Shi, S., Wang, B., Wang, G., Wang, W., Xu, D. Fault diagnosis and tolerant control of single IGBT open-circuit failure in modular multilevel converters (2016) IEEE Transactions on Power Electronics, 31 (4), art. no. 7153559, pp. 3165-3176.
• Rothenhagen, K., Fuchs, F.W. Performance of diagnosis methods for IGBT open circuit faults in voltage source active rectifiers (2004) PESC Record - IEEE Annual Power Electronics Specialists Conference, 6, pp. 4348-4354.
• Freire, N.M.A., Estima, J.O., Marques Cardoso, A.J. Open-circuit fault diagnosis in PMSG drives for wind turbine applications (2013) IEEE Transactions on Industrial Electronics, 60 (9), art. no. 6236145, pp. 3957-3967.
• Estima, J.O., Cardoso, A.J.M. A new approach for real-time multiple open-circuit fault diagnosis in voltage-source inverters (2011) IEEE Transactions on Industry Applications, 47 (6), art. no. 6022771, pp. 2487-2494.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper presents and test a novel diagnose for single and double IGBT modules open-circuit faults of grid-side converter for permanent magnet synchronous generator wind turbines. Although language and spell check is highly recommended the paper is well written and structured It provides interesting and relevant results. Following adjustments are suggested prior to publication:

- Abstract and other chapters: since failure state no. 1 corresponds to normal operation of the device confusion can be avoided if the paper clearly explains that the analysis relates to normal operation (state no. 1) and additional 21 failure states instead of mentioning 22 failure states only (e.g. by mentioning in line 21/22: “(…) with wind speed of 22 operational states including normal state and 21 failure states of PMSG (…)”).

- Line 37: not clear what is meant by “complex weather climate”. Does is stand for “weather conditions”?

- Line 72: Abbreviations BPNN and CART should be explained here as they appear for the first time in the paper (instead of Chapter 4).

- Line 80: please explain what kind of experiments are conducted in the paper.

- Chapter 2.2: please explain and refer in the text to the grouping/categories of the faut types as stated in Table 1.

- Line 142-144: Sentence “Assuming …” is not clear, please check grammar.

- Line 144-149: Use “First”, “Second, “Third”, instead of “1”, “2” and “3” in the text.

- Line 154: Please use full sentence explain how this step relates to the modeling procedure.

- Line 206-216: please re-work the description of DBM training model. In particular, Step 2 is missing.

- Line 218-224: explain the novel method as presented in Figure 3 in more detail. In particular, explain the sub-steps of the trend-feature analysis and DBN and classification results.

- Chapter 4: explain and refer to Table 2.

- Line 312: please check the grammar in sentence “Extracting …”

The writing of the paper needs improvement. The paper includes some grammatical errors as well as several typos, e.g.: Line 43: “too” instead of “to”; Line 79: ”deal” instead of “deals”; line 80. “simulation”; line 88 “equipped with”; line 104: “principles”; line 134: “2 corresponds to 000001”

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

I am satisfied with the present (revised and improved) version of the paper.