Sparse Identification of Nonlinear Dynamic Models in Power Systems
Grant Agreement No.: BPN/NWP/2025/1/00009
Granting Authority: Polish National Agency for Academic Exchange (NAWA), Scientists in Need Programme.
Project Duration: 1 February 2026 – 31 January 2028 (24 months).
Project’s Grant Amount: PLN 436,800.00
Grant Amount for WUT: PLN 436,800.00
Project Manager at WUT: Dr. Eng. Jakub Możaryn
Project Team:
- Dr. Leila Saeidiasl – researcher responsible for conducting the scientific research.
- Dr. Eng. Jakub Możaryn – scientific supervisor and project coordinator at Warsaw University of Technology.
- Participation of students and early-career researchers in research activities is also envisaged.
Project Description
The project aims to develop novel methods for the sparse identification of nonlinear dynamical systems, with particular emphasis on preserving the stability of power systems. The research focuses on advancing Sparse Identification of Nonlinear Dynamics (SINDy) methods by integrating data-driven modelling with Lyapunov stability theory, mathematical optimisation, and physics-based constraints.
The project will develop identification algorithms incorporating stability constraints, bilevel optimisation methods, and techniques for analysing model identifiability and robustness against disturbances. The proposed approaches will be validated using benchmark power system models and measurement data obtained from Phasor Measurement Units (PMUs).
The expected outcomes include advances in mathematical methods for identifying accurate, interpretable, and stable dynamical models, as well as the development of open-source software in Python and Julia. The resulting methods may support power system stability monitoring, early detection of potentially unstable operating conditions, and the design of advanced control systems.
The project will also contribute to strengthening international scientific cooperation and enhancing the research capacity of the Faculty of Mechatronics at Warsaw University of Technology.
Funding Acknowledgement
This project is funded by the Polish National Agency for Academic Exchange (NAWA) under the Scientists in Need Programme, project No. BPN/NWP/2025/1/00009.
