Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Data-driven control of renewable energy systems in grid-forming mode
Do you want to help the world achieve 100% renewable energy generation and get rid of fossil fuels? Are you interested in researching on advanced control techniques for renewable energy generators? If so, you are looking at the right project! In this project, we will focus on two emerging and promising control techniques that are essential for achieving 100% renewable generation in modern power grids, i.e., grid-forming control and data-enabled predictive control (DeePC). It has been well-acknowledged that grid-forming control is the key for 100% renewable generation as conventional control schemes relies on active sources in the grid (e.g., thermal power units). However, existing grid-forming control methods are not adaptive to changes in power grids and may have poor performance under certain situations. In this project, we will address this issue by using a data-driven approach (DeePC).
The massive integration of renewable generators significantly changes the characteristics of modern power systems. Conventional control of renewable energy is model-based, which can hardly lead to optimal performance in reality because the power grid is ever-changing. Hence, data-driven approaches are preferred as it utilizes the great availability of input/output data to capture the essential dynamics of the power grid, thereby avoiding accurate and time-consuming modeling for the power grid. In this project, we will explore the application of a novel data-enabled predictive control (DeePC) algorithm in renewable generation systems to perform safe and optimal control in a data-driven fashion. The DeePC algorithm is a model-free approach for calculating optimal control inputs solely based on input/output data, which provides the possibility to achieve network synchronization and power regulation of renewable generators in a data-driven fashion. We will investigate the role of the control objective function in determining the behavior of the renewable generator, and analyze how to specify a proper control objective function to make the generator operate in grid-forming mode. We will compare the data-driven approach with the state-of-the-art and investigate how DeePC facilitates a flexible and reliable grid-forming mode. We will also explore how the grid-forming mode with DeePC achieves a flexible and reliable 100% renewable power grid.
The massive integration of renewable generators significantly changes the characteristics of modern power systems. Conventional control of renewable energy is model-based, which can hardly lead to optimal performance in reality because the power grid is ever-changing. Hence, data-driven approaches are preferred as it utilizes the great availability of input/output data to capture the essential dynamics of the power grid, thereby avoiding accurate and time-consuming modeling for the power grid. In this project, we will explore the application of a novel data-enabled predictive control (DeePC) algorithm in renewable generation systems to perform safe and optimal control in a data-driven fashion. The DeePC algorithm is a model-free approach for calculating optimal control inputs solely based on input/output data, which provides the possibility to achieve network synchronization and power regulation of renewable generators in a data-driven fashion. We will investigate the role of the control objective function in determining the behavior of the renewable generator, and analyze how to specify a proper control objective function to make the generator operate in grid-forming mode. We will compare the data-driven approach with the state-of-the-art and investigate how DeePC facilitates a flexible and reliable grid-forming mode. We will also explore how the grid-forming mode with DeePC achieves a flexible and reliable 100% renewable power grid.
1. The student will be introduced to and taught in the concept of power converter (renewable generation) controls and data-enabled predictive control.
2. The student will be introduced to Simulink-based modelling and simulation of power converters.
3. The student will possibly have access to real power-electronic converter platform (if she/he is interested).
4. The student will implement the data-driven controller (DeePC) in simulations (and possibly experiments).
5. The student will compare the performance of the controller to the state of the art.
6. The student will investigate the design principle of control objective functions to facilitate the operation of grid-forming mode.
7. The student will write the report and prepare a presentation.
Corona Disclaimer:
This project could be done in person at the Automatic Control Laboratory, hybrid, or completely remotely depending on the current ETH rules. Most importantly, we can change between these forms whenever needed.
The project can be adapted on the run if new interesting research directions arise.
Finally, if the results are promising, they can be turned into a publication.
1. The student will be introduced to and taught in the concept of power converter (renewable generation) controls and data-enabled predictive control. 2. The student will be introduced to Simulink-based modelling and simulation of power converters. 3. The student will possibly have access to real power-electronic converter platform (if she/he is interested). 4. The student will implement the data-driven controller (DeePC) in simulations (and possibly experiments). 5. The student will compare the performance of the controller to the state of the art. 6. The student will investigate the design principle of control objective functions to facilitate the operation of grid-forming mode. 7. The student will write the report and prepare a presentation.
Corona Disclaimer: This project could be done in person at the Automatic Control Laboratory, hybrid, or completely remotely depending on the current ETH rules. Most importantly, we can change between these forms whenever needed. The project can be adapted on the run if new interesting research directions arise. Finally, if the results are promising, they can be turned into a publication.