the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 29 Sep 2025
Condition monitoring of wind turbine drivetrains: State-of-the-art technologies, recent trends, and future outlook
Abstract. As wind energy scales up to meet global decarbonization and energy security goals, reducing the LCoE has become essential, particularly through improvements in O&M. This positioning paper explores the state-of-the-art in condition monitoring for wind turbines. It focuses on drivetrain components, which are among the most failure-prone and maintenance-intensive subsystems. It examines current diagnostic and prognostic strategies using SCADA data, vibration and acoustic analysis, and digital twin frameworks, alongside emerging techniques in machine learning, signal processing, and hybrid modelling. The paper also identifies key challenges, including data availability, labelling, standardization, and the gap between academic research and industrial adoption. This work aims to guide future research and industrial efforts in making wind energy more reliable, predictable, and cost-efficient.
Competing interests: Yolanda Vidal, Yi Guo, Amir R. Nejad, and Shawn Sheng are members of the editorial board of Wind Energy Science.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(27163 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on wes-2025-168', Anonymous Referee #1, 06 Nov 2025
This is a very good, wide-ranging position paper on the subject. The group of distinguished authors have done an excellent job of coordinating such an informative work. It will be of interest to readers of the journal and should be published. Because it is so broad in scope, it is rather long. However, it is hard to see how any individual sections could be reduced. It might perhaps have been better published as a book, but if the journal board is happy with this format, so am I.It can be published as it is. I have suggested some minor points that may be of help to the authors.Section 1.1, p.2 – Consider starting with the formal definition of a drivetrain (it does appear later, so this is a minor point).Section 1.2.2 – A few references here would be useful.Section 1.2.2 – It would make sense to list these either in order of importance or in the order they appear in the drive train.Section 1.3.1 – How about presenting these levels as bullet points for quicker reading?L216–219 – It would be helpful to include a sentence giving examples of what these risks might be.L225 – Remove “etc.”L250–260 – This paragraph feels somewhat out of step with the paper. It is not clear what “Task 43” (and the later “second phase”) refers to, or what the organisational body is. This paragraph reads more as a description of task objectives and does not fit the overall narrative of the paper, which is about status.Section 3.1.1 – This is a comprehensive list of ML methodologies, but it lacks a clear takeaway message — e.g., which tools are best or most appropriate, or some guidelines for the operator. Table 1 is good and visual; could some of the extensive text in Sections 3.1.2 and 3.1.3 be replaced with a similarly well-crafted table?L727 – “This section has summarised…” use past tenseL731 – The phrase “according to the failure mode” — is it clear in the preceding section which steps are required for which expected failure mode?L1076, L1133 – Use the SCADA acronym, as it is now well established in the paper.L1425 – Autarkic is an uncommon word; I had to look it up. According to Wikipedia, autarky means self-sufficiency, usually applied to societies, communities, states, or economies. Is this the correct word in this context?L1462–1466 – The paper should report only work or concepts that have already been achieved, not future plans.Section 5.2.2 – CMS Signal Processing – The unnumbered sections on SAW and TFM do not fit well under this title (they are also not unique to gearboxes, which is the focus of Section 5.2). A lot of space is dedicated to SAW and TFM, which are two new sensor concepts, but very little is given to oil debris monitoring, which is well established. This feels out of balance.L1648 two periods ..Citation: https://doi.org/
10.5194/wes-2025-168-RC1 -
RC2: 'Comment on wes-2025-168', Anonymous Referee #2, 28 Jan 2026
1. Several important statements would benefit from stronger and more specific literature support. For example, the sentence “The emergence of floating offshore wind turbines (FOWTs) further introduces platform-induced motions and additional dynamic loads, amplifying the operational stresses on critical components” should be supported by relevant and up-to-date references that explicitly quantify or demonstrate these effects.
2. In Section 1.2.2, I recommend restructuring the discussion. It would be clearer to first summarize the main failing components and dominant failure mechanisms that are common across wind turbines in general, and then explicitly highlight the additional or emerging challenges specific to offshore and floating offshore wind turbines (e.g., harsher environments, coupled dynamics, accessibility constraints).
3. While the manuscript provides a comprehensive overview of existing condition monitoring (CM) techniques, the discussion remains largely descriptive. The authors are encouraged to more clearly articulate the methodological novelty of this review and to emphasize recent or emerging CM paradigms rather than primarily consolidating established approaches.
4. The current structure follows a technology- and method-centric organization, which makes it difficult to identify the underlying review logic. Reorganizing the manuscript around key CM problems (e.g., fault detection, diagnosis, prognosis, and decision-making) or fundamental challenges (e.g., data scarcity, generalization, interpretability) could significantly improve clarity.
5. Section 1.3 provides a solid conceptual overview of digital twins; however, it remains largely at the framework and terminology level. A more critical discussion comparing how different modelling strategies (data-driven, physics-based, and hybrid) actually improve CM performance would strengthen this section.
6. The SCADA-based CM section is extensive, but it does not sufficiently discuss fundamental limitations such as limited observability of early-stage mechanical faults, temporal resolution constraints, and confounding environmental effects. Explicitly addressing what SCADA-based methods cannot reliably detect would add critical depth.
7. While prognostics and remaining useful life (RUL) estimation are mentioned, the manuscript does not adequately analyze why these methods are still rarely adopted in industry. A discussion of challenges such as uncertainty quantification, validation difficulty, and integration with maintenance decision-making is recommended.
8. Although offshore and floating turbines are referenced throughout the manuscript, their CM challenges are not systematically contrasted with onshore turbines. A clearer and more explicit comparison would help readers understand what genuinely changes for CM when moving from onshore to offshore and floating systems.
Citation: https://doi.org/10.5194/wes-2025-168-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 619 | 1,201 | 30 | 1,850 | 45 | 53 |
- HTML: 619
- PDF: 1,201
- XML: 30
- Total: 1,850
- BibTeX: 45
- EndNote: 53
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1