The paper presents a systematic approach to developing Physics-Informed neuro-Models (PIM) for the transient analysis of power grids interconnected with renewables. PIM verifies itself as an adequate digital twin of power components, taking full advantage of physical constraints while requiring only a small fraction of data for training. Three new contributions are presented: 1) An PINN-enabled neuro-modeling approach is devised to construct an accurate EMT model; 2) A data-physics hybrid learning approach is substantiated to demonstrate its data efficiency and adaptability at various levels of data; 3) A balanced-adaptive PIM exemplifies its applicability when multiple terms are predicted simultaneously while maintaining alignment with physical principles. Under various operational scenarios, tests on rotating and static electric components as well as an IEEE test system verifies the efficacy and efficiency of the new learning-infused transient grid analytics.
No takes yet. Share an insight, caveat, or question.
Shen et al. (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: