Identification of Forced Oscillation Sources in Wind Farms using E-SINDy

Abstract

The rapid growth of wind power generation has led to increased interest in understanding and mitigating the adverse effects of wind turbine wakes and forced oscillations in wind farms. In this paper, we model a wind farm consisting of three wind turbines connected to a distribution system. Forced oscillations due to wind shear and tower shadow are injected into the system. If these oscillations are unchecked, they could pose a severe threat to the operation of the system and damage to the equipment. Identifying the source and frequency of forced oscillations in wind farms from measurement data is challenging. Thus, we propose a data-driven approach that discovers the underlying equations governing a nonlinear dynamical system from measured data using the Ensemble-Sparse Identification of Nonlinear Dynamics (E-SINDy) method. The results suggest that E-SINDy is a valuable tool for identifying sources of forced oscillations in wind farms and could facilitate the development of suitable control strategies to mitigate their negative impacts.

References in corpus (4)

• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data
• Ensemble-SINDy: Robust sparse model discovery in the low-data, high-noise limit, with active learning and control
• Identification of Surface Defects on Solar PV Panels and Wind Turbine Blades using Attention based Deep Learning Model