Voltage Sag Economic Loss Assessment Method via Physics-Informed Non-equilibrium Dynamics with Minimal Data Requirements

Voltage sag has emerged as a critical power quality issue causing substantial economic losses to sensitive industrial processes. Existing assessment methods suffer from either limited accuracy due to single-factor considerations or impractical data requirements for comprehensive approaches. This paper presents a minimal data requirement modeling framework based on non-equilibrium thermodynamics theory to overcome these limitations. The proposed method establishes a dynamic model that captures the intrinsic evolutionary mechanisms of voltage-sensitive processes. Through rapid model calibration with minimal measurement data, the approach reduces data requirements from extensive long-term monitoring to sparse sampling at critical points, eliminating engineering barriers of large-scale field experiments. Transforming uncertain voltage tolerance curve (VTC) regions into deterministic events via process parameter deviation quantification, dramatically enhancing practical applicability. Validation demonstrates over 90% accuracy across diverse sag events, providing quantitative basis for optimal mitigation equipment deployment and investment decisions.