Statistical physics approach to energy time series: Physics-inspired modeling meets stochastic functionals

• Applicant: Dr. Cai Dieball
• Subject Area: Statistical Physics, Nonlinear Dynamics, Complex Systems, Soft and Fluid Matter, Biological Physics
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 570237743

Stochastic fluctuations play a crucial role in many complex systems. In particular, they represent a key challenge in the understanding and modeling of energy dynamics, such as frequency and voltage fluctuations in electrical grids, or power generation and consumption in various systems. The transition towards renewable energy sources, that are inherently stochastic on many timescales due to changing weather, amplifies the fluctuations and consequently the risks endangering the stable operation of energy systems. This necessitates an increasing amount of expensive control measures, which, in turn, makes a profound understanding and modeling of the involved stochastic processes indispensable. In this project, we connect recent developments in the theory of stochastic processes with the immediate challenges in energy systems design and operation. We will advance the analysis and modeling of energy time series by refining physics-inspired approaches involving stochastic differential equations and machine learning. Motivated by recent progress in the fields of anomalous diffusion and stochastic thermodynamics, we will revisit the non-Gaussian short-time fluctuations of energy time series. Most importantly, we will introduce stochastic functionals in the form of time-integrated observables and first-passage times to better understand intricate features of the complex dynamics and the drivers of extreme events.

• DFG Programme: WBP Position