Appel à candidatures, Recrutement | Recherche, Emploi

PhD position onImpacts of Atmospheric Turbulence on Wind Turbine Main-Bearing Function and Failures, Univ. of Colorado (Boulder, USA)

Du 1 octobre 2020 au 30 septembre 2023

Site actualite

Laboratory: National Renewable Energy Laboratory (NREL) in Boulder Colorado, USA


  1. ● Dr. Edward Hart at the University of Strathclyde (
    ● Professor James Brasseur at the University of Colorado Boulder (


We seek an exceptional PhD candidate with the relevant background, motivation and ability to undertake a challenging, high-performance computing fluid-dynamics oriented programme of research within the Wind & Marine Energy Systems & Structures comprehensive doctoral training programme (CDT)* at the University of Strathclyde (Glasgow, UK).

 page1image18796400 page1image18794320 page1image18794944page1image18794736
Impacts of Atmospheric Turbulence on Wind Turbine Main-Bearing Function and Failures

PhD student sought for an exciting and challenging wind energy project with strong international collaborations in the UK and USA.



We seek an exceptional PhD candidate with the relevant background, motivation and ability to undertake a challenging, high-performance computing fluid-dynamics oriented programme of research within the Wind & Marine Energy Systems & Structures comprehensive doctoral training programme (CDT)* at the University of Strathclyde (Glasgow, UK). At least two years of the student’s research will take place with project partners at the University of Colorado and the National Renewable Energy Laboratory (NREL) in Boulder Colorado, USA. Key information is listed below, followed by an in-depth description of the proposed research on subsequent pages. This project is fully funded, with the successful applicant receiving an enhanced EPSRC stipend for 4 years. Applicants should be UK or EU nationals or have been domiciled in the UK for significant time, if in doubt please get in touch to discuss eligibility. The start date is planned for October 2020, with allowances to be made for the current Covid-19 situation. * tures/

Prerequisites: we seek a student with the following qualifications:

  • ●  Strong background and expertise in graduate-level fluid dynamics and aerodynamics

  • ●  Experience in computational fluid dynamics and high-performance computing

  • ●  Experience in computer code development

  • ●  A strong interest in interdisciplinary learning and research, computational methods, mathematical modeling and mechanics-based analysis

  • ●  Interest in international collaborations with a willingness to live in Boulder Colorado and Glasgow Scotland for roughly two years in each location (travel support provided)

  • ●  An ability to work well in a team and collaborate across a large project with multiple ongoing sub- projects.

    Please contact either of these individuals below for more information:
    ● Dr. Edward Hart at the University of Strathclyde (
    ● Professor James Brasseur at the University of Colorado Boulder (

    To apply: In the first instance please submit your complete CV, full university transcript (including all classes and grades) and a cover letter outlining why you are keen to work on this project and why you feel you are qualified and able to successfully undertake research of this nature.

PhD Project Proposal
Title: Impacts of Atmospheric Turbulence on Wind Turbine Main Bearing Function and Failures

First Supervisor: Dr. Edward Hart (University of Strathclyde).
Second Supervisor: Prof. James Brasseur (University of Colorado Boulder).
Industrial Partner: Dr. Jonathan Keller, National Renewal Energy Laboratory (NREL), Boulder, CO, USA.

Technical Background and Hypotheses

The continued expansion of wind energy within nations’ energy portfolios requires continued reductions in the levelized cost of energy (LCOE), the ratio of financial cost to purchase and operate wind farms to financial gain from the electrical power produced by the wind farm. Major contributors to the numerator include replacement costs for premature component failure on the drivetrain, including, in particular, the main bearing (Hart et al. 2020a). Whereas main bearing failure is likely not subsurface fatigue-related (Hart, et al. 2019) and a number of potential mechanisms likely contribute, the dominant processes underlying premature main bearing failures are not currently known (Hart, et al. 2019,2020a). Work has been undertaken to develop a systematic approach to the study of main bearing loading and failure mechanisms (Hart 2020b); wherein, it is demonstrated that a detailed understanding of the loads experienced by the main bearing, and their causal mechanisms, is a necessary prerequisite to progress in this field. The proposed research programme centers on the hypotheses that (1) the mechanisms underlying premature main bearing failures result from specific repetitive time changes in the bearing load-zone, and (2) that these deleterious load-zone forcings are in response to specific temporal characteristics in the moments and forces on the main shaft that result from the passage of the energy-dominant atmospheric turbulence eddies through the wind turbine rotor plane.

Utility-scale wind turbines and wind farms respond to turbulence within the “atmospheric boundary layer” (ABL), the 1-2 km region of the troposphere adjacent to the earth’s surface. During the day, the structure of the turbulence eddies transported within the ABL is driven by strong convection and strong shear, a coherent eddy structure that varies systematically with the global stability state of the ABL (Khanna & Brasseur 1998, Jayaraman & Brasseur 2014). As the eddies interact with rotating wind turbine blades, high temporal variabilities in aerodynamic loads pass from the rotor hub to the main shaft in the form of torque, bending moment and axial force fluctuations (Vijayakumar, et al. 2016) with three characteristic time scales (Nandi, et al. 2017) in main shaft moments (Lavely 2017). The shortest of these is below 1 sec. with relative variability on order 50% (Vijayakumar, et al. 2016). Lavely (2017) showed that, whereas main shaft torque fluctuations respond to rotor- averaged horizontal winds, shaft bending moment fluctuations respond to time changes in spatial asymmetry in horizontal velocity over the rotor plane, generated as turbulence eddies pass through the rotor.

If the above hypotheses are valid, it follows that the turbulence-generated deleterious load fluctuations on the main bearing are likely driven by different classes of turbulence structure and loading response at different time scales as atmospheric eddies pass through the rotor plane. Since ABL turbulence structure varies with atmospheric stability, deleterious load characteristics change during the day and among seasons, as well as with topography. Furthermore, within a wind farm the hypotheses can be extended to include potential deleterious main bearing responses to combinations of atmospheric and rotor wake turbulence eddying motions due especially to the generation of spatial asymmetries over wind turbine rotor planes.

Aims and Objectives

The overall objective of the proposed research programme is to characterize the impacts of energetic atmospheric turbulence eddies on the time variations in the loads on wind turbine main bearings using high- fidelity computer models and simulations. Our specific objectives include:

  1. Development of high-resolution large-eddy simulations (LES) of the daytime atmospheric boundary layer (ABL) at a minimum of two different stability states (University of Colorado Boulder).

  2. Development of advanced wind turbine blade actuator line models of turbine rotors embedded within the atmospheric boundary layer LES, including capabilities to simulate both rigid and elastic blade response (University of Colorado Boulder).

  3. Application, and possible extension, of contact models of the wind turbine main-bearing-on-main-shaft to investigate effects of ABL generated loading. The focus of this stage will be the characterization of spatio- temporal changes in main bearing load zones and response magnitudes in relation to time changes in main shaft applied moments and loading; in addition, load impacts for individual rollers and the possible damage resulting will also be studied (University of Strathclyde).

  4. Development, execution and analysis of computational experiments to generate large datasets for quantifications and analysis of main bearing response to the passage of atmospheric and wake turbulence eddies through wind turbine rotors in context with the hypotheses above.

Location of Research Efforts

The first two specific objectives will be developed at the University of Colorado Boulder under the direct supervision of Professor Brasseur and with input from Dr. Keller at NREL and Dr. Hart at Strathclyde. These will be integrated with specific objective 3 to be developed at Strathclyde University under the direct supervision of Dr. Hart. Analysis and all developments will be overseen by the Strathclyde-Colorado-NREL team.


As summarized in Aims and Objectives, we propose to develop large-eddy simulations (LES) of the atmospheric boundary layer in which are embedded wind turbine rotors using actuator line models (ALM) for the wind turbine blades. This is conceptually similar to the LES/ALM model developed in the PhD thesis of Adam Lavely (Professor Brasseur’s former student at the Pennsylvania State University) who studied the interactions between a single 5 MW wind turbine and the turbulent eddy structure of the daytime atmospheric surface layer. Through the collaboration with Strathclyde, we propose to incorporate different levels of main bearing models, including quasi-static (Hart 2020b) and dynamic model representations of roller forces generated by loadings on the low-speed shaft driven by nonsteady aerodynamic forcing of the wind turbine blades and rotor from the passage of specific classes of atmospheric turbulence eddies.

Project Plan

As described in Technical Background and Hypotheses, nonsteady loadings on the drivetrain include responses to aerodynamic forcing from atmospheric and wake turbulence. As summarized in Aims and Objectives, the project plan is to apply computer models and computer simulation to quantify the temporal changes in the main bearing load zone in response to the asymmetrical passage of atmospheric turbulence eddies through wind turbine rotors with the consequent creation of temporal changes in main shaft moments and forces. High-fidelity large-eddy simulation of the atmospheric boundary layer (ABL) at two disparate stability states together with the inclusion of actuator line rotor models of wind turbine rotors will allow for the characterization of differences in load response with different atmospheric turbulence eddy structure. Comparisons between computational experiments that describe the response of first rigid and then elastic wind turbine blade models to the passage of ABL turbulence through the wind turbine rotor plane will allow for model complexity to be increased gradually – ensuring proper verification of implementations – while also allowing for a quantification of the relative impacts of elasticity on drivetrain loading. This therefore broadens the applicability of the project to also inform as to possible effects experienced by larger turbines, which tend to have increased blade flexibility. Furthermore, within the more complex models of the main bearing, we aim to include comparisons of the spherical vs. tapered roller bearing (SRB vs TRB) in order to expand the applicability of the study to main bearing design.

Applicant Prerequisites

This will be a challenging PhD project on a number of levels and hence requires a researcher of outstanding technical ability who is driven, highly motivated, enjoys learning, and wants to be a part of a world leading research programme. A strong mathematical background is necessary in addition to fluid dynamics training up to a graduate level. Computer programming and complex code development will also be key skills throughout the research programme. Applicants should be interested in challenging multidisciplinary research and willing to put significant time into self-study and guided-study on necessary topics as they arise. An ability to communicate effectively with all partners in an international setting will also be important. Finally, this project will require significant time to be spent in both Scotland and the US and hence applicants must be prepared to spend roughly 2 years in each location, first in Boulder, Colorado and then Glasgow, UK.


Hart, E., Turnbull, A., Feuchtwang, J., McMillan, D., Golysheva, E., Elliott, R. 2019 Wind turbine main-bearing loading and wind field characteristics. Wind Energy. 2019;1–14.

Hart, E., Clarke, B., Nicholas, G., Kazemi Amiri, A., Stirling, J., Carroll, J., Dwyer-Joyce, R., McDonald, A., Long, H. 2020a A review of wind turbine main-bearings: design, operation, modelling, damage mechanisms and fault detection. Wind Energ. Sci., 5, 105–124,

Hart, E. 2020b Developing a systematic approach to the analysis of time-varying main-bearing loads for wind turbines. Under review with Wiley Wind Energy. Pre-print available on request.

Jayaraman, B., Brasseur, J. 2014 Transition in Atmospheric Turbulence Structure from Neutral to Convective Stability States, 32nd ASME Wind Energy Symposium, 13-17 Jan. 2014, National Harbor, MD, AIAA 2014-0868.

Khanna, S., Brasseur, J.G. 1998 Three-dimensional buoyancy and shear induced local structure of the atmospheric boundary layer, J. Atmospheric Sciences, vol. 55 (5): 710–743.

Lavely, A. 2017 Effects of Daytime Atmospheric Boundary Layer Turbulence on The Generation of Nonsteady Wind Turbine Loadings and Predictive Accuracy of Lower Order Models, PhD Thesis, Penn State University.

Nandi, T.N., Herrig, A., Brasseur, J.G. 2017 Nonsteady wind turbine response to daytime atmospheric turbulence. Phil. Trans. R. Soc. A 375: 20160103..

Vijayakumar, G., Brasseur, J.G., Lavely, A., Jayaraman, B., Craven, B.C. 2016 Interaction of Atmospheric Turbulence with Blade Boundary Layer Dynamics on a 5MW Wind Turbine using Blade-Boundary-Layer- Resolved CFD with hybrid URANS-LES. AIAA paper 2016-0521, AIAA SciTech 2016, San Diego, CA.