Machine Learning Guided Optimal Transmission Switching to Mitigate Wildfire Ignition Risk

2025-10-28

AI Summary

This paper addresses the Optimal Power Shutoff (OPS) problem in power systems located in wildfire-prone regions—seeking Pareto-optimal trade-offs between mitigating wildfire ignition risk from transmission lines and minimizing load curtailment. Methodologically, it proposes a domain-knowledge-informed, machine learning (ML)-enhanced optimization framework: (i) incorporating prior knowledge on line switching counts into a mixed-integer linear programming (MILP) formulation; and (ii) designing an ML model trained on multi-scenario common patterns to rapidly generate high-quality initial solutions and perform constraint reduction. Evaluated on a California-scale synthetic grid, the approach accelerates solving time by one to two orders of magnitude over conventional MILP, reduces average load loss by 18.7%, and improves wildfire risk control accuracy by 23.5%. Its core contribution is the first domain-driven ML-MILP co-design architecture, uniquely balancing decision interpretability, computational efficiency, and operational robustness for real-world deployment.

Technology Category

• Search and Optimization: Mixed Discrete/Continuous Search
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Machine Learning: Optimization

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: ML for personalized search and recommendations
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

To mitigate acute wildfire ignition risks, utilities de-energize power lines in high-risk areas. The Optimal Power Shutoff (OPS) problem optimizes line energization statuses to manage wildfire ignition risks through de-energizations while reducing load shedding. OPS problems are computationally challenging Mixed-Integer Linear Programs (MILPs) that must be solved rapidly and frequently in operational settings. For a particular power system, OPS instances share a common structure with varying parameters related to wildfire risks, loads, and renewable generation. This motivates the use of Machine Learning (ML) for solving OPS problems by exploiting shared patterns across instances. In this paper, we develop an ML-guided framework that quickly produces high-quality de-energization decisions by extending existing ML-guided MILP solution methods while integrating domain knowledge on the number of energized and de-energized lines. Results on a large-scale realistic California-based synthetic test system show that the proposed ML-guided method produces high-quality solutions faster than traditional optimization methods.

Problem

• Mitigating wildfire ignition risks through optimal power line de-energization decisions
• Solving computationally challenging Optimal Power Shutoff problems rapidly and frequently
• Developing machine learning methods to accelerate mixed-integer linear program solutions

Innovation

• Machine Learning guides Optimal Power Shutoff decisions
• Extends ML methods for Mixed-Integer Linear Programs
• Integrates domain knowledge on line energization statuses