Project Start Date : Start of the next semester
Application Deadline : Open Until Filled
Description:
Significant power losses due to complex interactions of wind turbine wakes in wind farms call for the development of novel wake control strategies. Yaw angle control is considered to be an effective strategy by deflecting the wakes away from downwind turbines. However, both experimental and numerical studies are inappropriate for power optimization purposes, and computationally-cheap analytical-based models are demanded.
In this project, the goal is to develop a novel and efficient optimization tool to design the yaw angle control strategies for wind farm to improve the power output. A analytical-based models to predict the velocity field will be developed and a Bayesian optimization algorithm will be used to obtain the optimal yaw angle distribution. You will collaborate with Dr. Xiao Hu ([email protected]) to accomplish the following two tasks:
- Combine the wake superposition principle with the established point vortex model for wake prediction ( Zong & F. Porté-Agel, JFM, 2020) to generate velocity field.
- Develop a learning-based Bayesian optimization framework to identify the optimal yaw distribution.
We offer you:
- Guidance in developing the analytical-based models for wind farm.
- Bayesian optimization algorithm using Gaussian processes.
Your qualities:
- Good data processing skills with Python or Matlab.
- Demonstrable interest in machine learning and wind energy optimization.
- Background in optimization and fluid mechanics will be a plus.
Expected outcomes:
- A strong understanding of the wake in a wind farm.
- Hands-on experience with the optimization tools.
- Development of academic and technical writing skills.
- Submission of a paper to a renowned international journal.
For more information about the project and to be considered the position, please contact:
Prof. Dr. Fernando Porte-Agel
Email: [email protected]
Dr. Xiao Hu
Email: [email protected]
Wind Engineering and Renewable Energy Laboratory