DC 8- Tensor optimization for storage integration

Project Title: Tensor optimization for storage integration
Advisors: C. Bordin (UiT), C. Riener (UiT), M. Schweighofer (UKON), Mentors: Michaël Gabay (Artelys)
Objectives: Optimal integration of storage in power and energy networks with a high share of intermittent renewable energy sources is one of the major challenges for a successful transition of decentralized energy production. The goal is to optimally manage the choice between using energy “now” and saving it in the store to support future system adequacy. The resulting On one hand, mixed-integer linear and non-linear modeling has been successfully used to design energy and power networks. On the other hand, conic optimization implementing carefully chosen fragments of non-linear formulations of the given problems via semi-definite programs have recently emerged as a powerful tool for designing tractable algorithms for power system operation. This project will bring together both techniques and further develop them with more general tensor based optimization techniques to study novel approaches to model key technological features of different types of storage technologies (namely battery degradation, seasonality of hydrogen,and efficiency of pumped thermal electricity storage) for their inclusion within mathematical optimization models.
Expected Results: Measure the value of non-linear optimization and semidefinite relaxation for the technological modeling of energy storage within power networks. Implementing these approaches with Artelys, France. Writing three scientific journal articles.
Planned secondment(s): 9 month secondment with Markus Schweighofer (M21-30) to formulate non-linear optimization approaches for pumped thermal electricity storage and 3 months with Michaël Gabay, Artelys (M34-36).
Joint degree: UiT The Arctic University of Norway, University of Konstanz

Comments are closed.