Here you will find job advertisements from the wind energy sector. Please use the input form to publish job advertisements.
Primary supervisor: Prof. Jan Helsen
The VUB Acoustics and Vibrations Research group and VUB AI-group work closely together in the field of machine monitoring. Novel signal processing and AI methods are developed specifically targeted at the prediction of failures and accurate assessment of their progression. In this context we work closely together with leading companies: Atlas Copco, BASF, ZF, … The team has a core focus on wind energy in the context of OWI-lab. There we have ongoing research projects with MHIVOW, ZF Wind Power, Parkwind, … Our multi-disciplinary approach allows us to bring methodological advancements all the way to application in industry.
The process of tracking the health of machinery is commonly known as condition monitoring. Typically, it involves recording data, analyzing this data, and then inspecting the resulting indicators for potential significant changes that could be symptomatic of a defect. Incorporating condition monitoring in the Operations and Maintenance of a company opens the door for predictive maintenance. At VUB we can offer help to companies in this condition monitoring process by performing specialized data analysis of their machines. This can be through the use of vibrations, rotation speed, acoustics, or other sources of measurable information. All these measurements typically produce a lot of complex data, therefore we investigate new ways how we can effectively and efficiently analyze this data to provide an as accurate as possible health summary of the machine. Next to data analysis, there is thus also a strong focus on big data processing, automation of the result interpretation using machine learning, and keeping up with the Internet of Things trend of increased sensorization and data acquisition.
PHD Project Description
The research focuses on developing new data analysis tools for condition monitoring of wind turbines and rotating machinery in general. The work will include implementing existing concepts in code, but also developing novel ideas for signal processing. There is a strong emphasis on bearing and gear monitoring. In addition to the development of novel methodologies for signal analysis, we also strive to deliver actionable information, relevant to the industry. Thanks to our strong connections with several industrial partners, we have the opportunity to work on interesting issues, but this means we also need to disseminate our results. Therefore, your work will go beyond the development of new methods and will also include expanding our data analysis platform with your new tools and combining your new tools with state-of-the-art machine learning approaches. The latter is accomplished by our collaboration with the Artificial Intelligence group of VUB. We offer the opportunity to work in a very inspired, motivated and enjoyable research group that is looking to expand. The focus is also not purely on academic aspects thanks to our industrial collaborations. Therefore, you will inevitably also gain significant industrial experience and insight into how companies function and how to operate together with them. On top of the meaningful academic and industrial experience that you will gain, we encourage every PhD student to go and present their work at international conferences abroad. A research stay of short and long duration at our partner institutions (NTNU, INSA, NREL,…) is also possible.
Applicants should preferably have:
- Master degree in Mechanical, Electrical, or Mathematical engineering
- A relevant Master’s degree and/or experience in one or more of the following will also be an advantage: wind turbine dynamics, signal processing, machine learning techniques, Bayesian statistics, ...
- Background or interest in programming (Matlab, python, java, C/C++, …)
- Proficiency in English is a plus
We offer an international open working environment stimulating personal development through international courses, many opportunities to attend and present at conferences abroad. Possibility to spend part of the research abroad. A competitive salary, public transport coverage and health insurance. The PhD normally lasts 4 years.
Successful candidates will be notified by September 15th 2019
HOW TO APPLY:
All applications should be made through e-mail (firstname.lastname@example.org)
Promoter: J. Meyers
Contact: Prof. J. Meyers, Department of Mechanical Engineering, Celestijnenlaan 300A, B3001 Leuven, Belgium. T: +32(0)16 322502. Google Scholar
Apply using the KU Leuven online application platform. (Applications by email are not considered!)
This PhD position is based in the Turbulent Flow Simulation and Optimization (TFSO) research group headed by Prof. Dr. Johan Meyers, which is part of the department of Mechanical Engineering. The position is embedded in a large interdepartmental project on “Efficient methods for large-scale PDE-constrained optimization in the presence of uncertainty and complex technological constraints”, which is funded by the special research fund of KU Leuven. The project is a collaboration between the research group of Prof. Meyers and the Numerical Analysis and Applied Mathematics Section (NUMA) of Prof. Vandewalle at the department of Computer Science. The PhD research is one of 10 research positions in this project, and focusses on developing efficient methodologies for optimal control of wind farms using multiphysics and multiscale wind-farm models, and their use for multi-objective power and load optimization.
In recent years there has been a lot of interest in control of wind farms to improve energy extraction or allow power tracking (relevant for power grid balancing and ancillary services). In most cases, control models that are considered are based on simplified wake models, and are often even steady state. KU Leuven pioneered the use of large-eddy simulation of the atmospheric boundary layer for optimal control of wind farms, considering not just steady state optimization of turbine set-points, but also dynamic changes of set-points that directly interact with the turbulent flow structures in the wind farm (see, e.g., Goit and Meyers, J. Fluid Mech. 768, 2015). This methodology has led to the discovery of new physical flow-control mechanisms for improved wake mixing based on dynamically induced vortex shedding (see, Munters & Meyers, Wind Energ. Sci. 3, 2018; Yilmaz & Meyers, Phys. Fluids 30, 2018). However, the practical realization of wind-farm control should not only incorporate power maximization or power tracking, but also aspects of unsteady loading, leading to a multi-objective optimal control problem. This requires an optimal control approach that includes both flow aspects as well as elements from nonlinear structural mechanics, i.e. an aero-elastic turbine model. Although such coupled approaches are available in high-fidelity wind-farm simulation environments such as SP-Wind (developed at KU Leuven), their integration in an optimal control framework is an open challenge. Such an integration opens up many interesting new perspectives, and in particular would allow for studying and understanding the interaction between turbulent flow structures, control and loads, which is a topic that eludes the wind-farm community to date.
Research: the research focuses on the formulation of efficient and fast optimal control methods for wind farm control based on coupled flow structural models of wind farms and wind turbines. This requires the development of accurate adjoint formulations for the nonlinear multibody turbine structural model in SP-Wind as well adjoint formulations for the multi-rate time stepping procedure that couples flow and structural models. Moreover, proper regularization approaches need to be developed for the objective gradients, as numerical errors in the flow model can contaminate the gradients in the structural model in the multiscale coupling approach. Finally, appropriate quasi-Newton or thrust-region optimization methods need to be developed that allow to escape local minima that are abundant in this type of optimization problems. Based on these new numerical methodologies, the research further focusses on the inclusion of fatigue loading in wind-farm optimal control studies, allowing for the first time to explore potential benevolent interaction between turbulence, control and structural loading of turbines in a wind farm.
Timeline and remuneration: Ideal start time is Fall 2019. The PhD position lasts for the duration of four years, and is carried out at the University of Leuven. During this time, the candidate also takes up a limited amount (approx. 10% of the time) of teaching activities. The remuneration is generous and is in line with the standard KU Leuven rates. It consists of a net monthly salary of about 2000 Euro.
Candidates have a master degree in one of the following or related fields: fluid mechanics, aerospace or mathematical engineering, numerical mathematics, or computational physics. They should have a good background or interest in fluid mechanics, optimization, simulation, and programming (Fortran, C/C++, Matlab, python, …). Proficiency in English is a requirement. The position adheres to the European policy of balanced ethnicity, age and gender. Both men and women are encouraged to apply.
To apply, use the KU Leuven online application platform (applications by email are not considered) Please include:
a) an academic CV and a PDF of your diplomas and transcript of course work and grades
b) a statement of research interests and career goals, indicating why you are interested in this position
c) a sample of technical writing, e.g. a paper with you as main author, or your bachelor or master thesis
d) two recommendation letters
d) a list of at least two additional references (different from recommendation letters): names, phone numbers, and email addresses
e) some proof of proficiency in English (e.g. language test results from TOEFL, IELTS, CAE, or CPE)
Please send your application as soon as possible and before October 1st, 2019 at the latest.
Decision: when a suitable candidate applies.
Starting date: candidates can start immediately. Start preferable in Fall 2019.