Postdoc AI-assisted combinatorial optimization for congestion management

To achieve the objectives of the Climate Agreement there is a need to rapidly increase the amount of energy that comes from solar and wind farms. In the Netherlands, there is significant interest in relevant investments. However, the lack of grid capacity in the medium voltage grid hinders the realization of new projects. Given the significant lead times and cost of medium voltage grid reinforcement, mobilizing flexibility that is available at the edges of the power system to increase the hosting capacity of the existing network by preventing or minimizing congestion is imperative.

To address this challenge, you will work with a consortium of industry and academic partners who aim to develop and validate scalable optimization and control techniques to exploit distributed energy storage assets located at the low voltage grid (residential and agricultural end-users) in order to resolve congestion issues at the medium voltage grid level and enable the seamless connection of large-scale solar and wind farms. In particular, you will explore the use of machine learning techniques in conjunction with mathematical programming to accelerate the solution of large-scale combinatorial optimization problems that are used to price and specify the parameters of congestion management instruments.

Your tasks will include:

• Developing multi-level distributed optimization and control algorithms for distributed energy storage devices.
• Designing and pricing new congestion management instruments that are compatible with current legislation.
• Validating the proposed algorithms via a pilot that takes place in a congested area where several large-scale renewable generation projects are in the pipeline and involves a significant amount of distributed storage.

You will primarily contribute to Intelligent Energy Systems research activities of the Electrical Energy Systems group. Besides research you will also have the opportunity to contribute to education within the department.

• PhD in Electrical Engineering, Operations Research, Computer Science or a related discipline with a focus on power systems and optimization/machine learning techniques. The position is also open to candidates at the final stage of their PhD who have not defended yet.
• Ability to conduct high quality academic research, demonstrated for instance by a relevant PhD thesis and/or publication(s).
• Proficiency in Python programming. Experience in collaborative software projects is desirable.
• Ability (or willingness) to teach and supervise students, shown by experience or assistance in teaching and positive evaluations of these teaching efforts.
• Excellent command of the English language and good communication skills. Note that there is no Dutch language requirement.
• Be a team player and able to work in a dynamic, interdisciplinary context.
• Be available to start no later than September 2024.

You must possess a comprehensive knowledge of power systems and/or electricity markets. You must have a proven ability to develop and apply quantitative decision-making methods (e.g., mathematical programming or data-driven/AI approaches), as evidenced by a relevant PhD thesis or publications. Furthermore, familiarity with topics such as congestion management, flexibility provision, energy management, energy communities, mathematical decomposition methods, distributed optimization techniques is highly desirable.