AI Could Help Power Companies Spot Grid Problems With Far Fewer Sensors
Fewer sensors, faster warnings — and maybe fewer blackouts for your neighborhood.
Researchers Farah Elsherif, Behrouz Azimian, and Anamitra Pal propose an integrated deep learning framework for simultaneous topology identification (TI) and distribution system state estimation (DSSE) in real-time unobservable primary distribution networks instrumented by a minimal set of synchronized measurement devices (SMDs). Their paper is accepted for presentation at the 2026 North American Power Symposium.
The problem: rapid integration of distributed energy resources is fundamentally altering power flow patterns in primary distribution networks and intensifying operational uncertainty, compounded by a lack of real-time situational awareness and frequent topology changes.
Their approach: a correlation-driven SMD placement algorithm is introduced first, jointly satisfying TI accuracy and DSSE performance requirements by exploiting temporal and spatial correlations in nodal voltage measurements. A dual deep neural network (DNN)-based DSSE model is then developed to estimate three-phase voltage magnitudes and angles across diverse operating conditions. To extend the framework beyond the base topology, fine-tuning-based transfer learning adapts the DSSE model to reconfigured topologies using limited retraining data.
The framework is validated under both Gaussian and non-Gaussian measurement noise and benchmarked against a conventional estimation approach and a single DNN model.
- The local power lines feeding your home are mostly unmonitored, so utilities often learn about problems only when customers call
- The AI estimates conditions for the whole network using just a few well-placed sensors, which could cut monitoring costs
- It adapts to grid changes quickly using transfer learning — retraining an existing AI instead of starting over
Why It Matters
Cheaper grid monitoring could mean fewer surprise outages, safer voltages for your appliances, and more rooftop solar allowed.