the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 23 Jan 2025
On reliability design and code calibration of wind turbine blade bearings under extreme wind conditions
Abstract. This study presents the reliability analysis of the blade bearing in the ultimate limit state. The National Renewable Energy Laboratory 5 MW reference wind turbine is selected for the study, and the Monte Carlo simulation is used for the reliability analysis and estimation of the probability of failure. The uncertainty in turbulence intensity as well as materials are considered in the reliability analysis. A sensitivity analysis is carried out to evaluate the effect of bearing dimension variation. It is observed that conformity and ball diameter have the most sensitivity in the dimension aspect of reliability. IEC standards, as well as wind conditions in different wind sites around the world are studied, and it is shown that the probability of failure in blade bearing is higher in most of the wind sites than in sites with IEC standard wind.
- Preprint
(1888 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 28 Feb 2025)
-
CC1: 'Comments on wes-2024-186', Matthias Stammler, 29 Jan 2025
reply
A very interesting work and a good approach to compare IEC classes with real sites!
This comment collects a few things I found during a first read, with a focus on practical aspects of blade bearings:Â
Line 13: "Blade bearings serve as the connection point between the rotor and the hub, allowing the blades to rotate around the hub. " In my understanding, the rotor consists of blades, blade bearings, and the hub. Thus I would rather say the blade bearings connect the blades with the hub. An I find it misleading to say the blade rotate around the hub. They rather rotate around their own primary axis?Â
Line 16: I would recommend to mention cage failures as well.Â
Line 20 to 30: There is also a publication by Schwack comparing the different RCF calculation methods.Â
Line 36 to 42: The plastic deformation of 0.0001D is set a limit value in the ISO76 and the 2009 version of the DG=3. However, the 2024 version of the DG03 opens possibilities to increase this value, mainly for two reasons:
- Pitch bearings in general can operate in raceway conditions that are considered a failure in other applications. Macroscopic spallings are not a reason to stop operating a pitch bearing and will not cause an exchange of it. Only when the risk of inoperability is imminent (i.e. expected friction torque too high for drive or loss of blade connection) an exchange is undertaken.Â
- In four-point contact ball bearings highest static loads in rolling contacts are at high contact angles. 'Normal' operation in power production is at lower contact angles, thus for the main part of its lift, the ball does not roll over the indent.Â
I would highly recommend to mention these concepts in your introduction as they heavily influence the general conclusions you possibly draw at the end. Also I would vey much recommend to get acquainted with the concept of damage and failure as described in the 2024 DG03 and use those terms consistently throughout the paper.Â
Figure 1 Step2: It should be "Blade root loads" instead of "Blade's root loads" I think, because the first is a common expression. Also it is a bit confusing because the sketch shows an airfoil used in the outer portions of the blade, but most certainly not a blade root. I would consider the lift at the blade root to be negligible.Â
Figure 1 Step3: Are you sure you obtain the load Q in N? Or should it be kN for this graph?Â
Figure 1 Step4: The sketch does not fit the caption. It shows to spherical bodies in unloaded contact, but certainly not a maximum Hertz stressÂ
Equation 7 please give a reference
Line 108: Capital Z instead of small zÂ
Section 3.2.1 Pitch bearing rings are commonly manufactured of 42CrMo4 steel - please elaborate on the choice of the studies on AISI51200 (100Cr6) steel property distributions - the hardening process is fundamentally different.
Section 3.2.2 While it is fundamentally true that all dimensions have a certain variance to them, this is somewhat countered by the assembly process: Rings and balls are matched to obtain a target friction torque in unloaded condition. Thus combining normal distributions for all parameters does not reflect reality.Â
Section 4.2.1 There is no such thing as very coarse machining of balls of this size. You usually buy them in batches of very fine tolerances and them match them to obtain target torques.Â
Section 4.2.3 Lower values of raceway conformity will drastically increase the friction torque and the likelihood of surface wear. it is questionable to use a value range this bigÂ
Section 4.2.4 It would make sense to differentiate between the nominal contact angle (as manufactured) and the 'contact angle' (changes as a function off load and ring deformation in operation). Currently, it is a bit unclear which one the authors refer to.ÂDisclaimer: this community comment is written by an individual and does not necessarily reflect the opinion of their employer.Citation: https://doi.org/10.5194/wes-2024-186-CC1
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 72 | 8 | 3 | 83 | 1 | 1 |
- HTML: 72
- PDF: 8
- XML: 3
- Total: 83
- BibTeX: 1
- EndNote: 1
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1