Event-Structured Physics-Informed Neural Networks for Differentiable Critical Clearing Boundaries

2026-07-30

AI Summary

This work addresses the challenge of accurately and efficiently estimating the critical clearing time (CCT) in transient stability assessment of power systems, where conventional methods rely on extensive repetitive simulations and struggle with complex fault scenarios. The authors propose an Event-Structured Physics-Informed Neural Network (ES-PINN) that aligns its architecture with the swing equations governing pre-fault, during-fault, and post-clearance phases, enforcing exact state continuity at event interfaces to eliminate discontinuity-induced errors. By integrating trajectory-informed stability margins, ES-PINN enables a differentiable approximation of the CCT boundary. A theoretical framework linking residuals, trajectories, and local CCT error is established to facilitate efficient stability boundary extraction and sensitivity analysis. Experiments on IEEE 9-, 14-, and 30-bus systems demonstrate that ES-PINN consistently outperforms baseline methods across diverse disturbances, achieving high accuracy with superior computational efficiency.

Abstract

Transient-stability assessment determines whether a power system can recover after a disturbance and is therefore essential to preventing generator trips and cascading outages. A key metric is the critical clearing time (CCT), which specifies the maximum time available to clear a fault before synchronism is lost. Reliable CCT estimation is challenging because complicated fault-clearing dynamics require repeated simulations over many fault severities and clearing times. We propose an event-structured physics-informed neural network (ES-PINN) that aligns its representation with the pre-fault, fault-on, and post-clearing swing dynamics and enforces exact state chaining across event interfaces. A smooth trajectory-induced stability margin defines a differentiable approximation of the CCT boundary, enabling accurate boundary extraction, local sensitivity analysis, and optional direct CCT prediction through a distilled readout. We further prove a local residual-to-trajectory-to-CCT error estimate, in which exact event chaining eliminates separate state-interface defect terms. Experiments on IEEE 9-, 14-, and 30-bus systems show that ES-PINN consistently improves held-out trajectory and stability-boundary accuracy over matched neural-surrogate baselines across mechanical and electrical contingencies with multiple clearing configurations. Additional full-network DAE validation, multi-fault experiments, and runtime analyses further demonstrate the effectiveness and computational efficiency of the proposed framework.

Problem

• transient stability
• critical clearing time
• power system dynamics
• stability boundary
• fault clearing

Innovation

• Event-structured PINN
• Critical Clearing Time
• Differentiable Stability Boundary
• Exact State Chaining
• Transient Stability Assessment