Resilience enhancement and value of information quantification for renewable-dominated power systems under seismic sequences via Bayesian perception

The transition to high-renewable power systems intensifies vulnerability to complex seismic hazards, yet existing resilience assessments often neglect the cumulative physical damage from seismic sequences and the economic potential of real-time situational awareness. This study proposes an integrated resilience framework that couples a Probabilistic Capacity Degradation Model with a Bayesian perception-informed recovery strategy. The Probabilistic Capacity Degradation Model effectively captures structural memory and nonlinear damage accumulation under Mainshock-Aftershock and Double-Mainshock events, while a novel Dynamic Spatio-temporal and Topological Coordinated Scheduling algorithm leverages Seismic Structural Health Monitoring data to support post-disaster restoration scheduling. Numerical simulations on a modified IEEE 118-bus system demonstrate that this perception-informed paradigm yields case-study-specific improvements in technical resilience compared to traditional manual inspection strategies. Under the adopted simulation assumptions, the proposed framework increases restoration speed by 62.5%, expands the comprehensive resilience envelope by 128.9%, and reduces the Loss of Load Probability by 46.7%. Furthermore, a life cycle cost-benefit analysis provides quantitative, case-study-specific economic evidence for sensing investments, yielding a Net Present Value of $359.5 million and a Benefit-Cost Ratio of 7.85. By explicitly quantifying the policy-conditioned economic value of information associated with sensing-enabled recovery, this research bridges the gap between engineering resilience assessment and financial decision-making, supporting evidence-informed prioritization of smart grid sensing infrastructure.