Abstract
Abstract In addressing the limitations of unclear mechanisms in data driven methods and the restricted applicability of inference interpolation methods in power data recovery research, this paper proposes a hybrid driven framework that combines both data driven and model driven approaches. This framework integrates data driven and model driven components through linear recombination, while explicitly demonstrating the interpretability of the model. By fitting three nonlinear transformation processes and comparing effective and ineffective mechanisms, the efficacy of model driven enhancement in improving data driven performance is validated. Furthermore, the effectiveness of the proposed framework is demonstrated through a data recovery task in IEEE-39 node system.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call