End User
SWW
Description
The AI-Based Network Model Calibration service implements a data-driven calibration process that links real PQM measurement data to a PowerFactory model and iteratively adjusts uncertain profile parameters until deviations between simulation and measurement are minimised. The service accounts for limited observability by applying direct measurement-driven calibration where sensors exist and structured prior knowledge where they do not, preventing overfitting while ensuring physically consistent estimates across the network.
Core Capabilities
Monitoring & Anomaly Detection
Predictive & Prescriptive Analytics
Business Need
Poorly parameterised network models propagate inaccuracies into all downstream model-based services — state estimation, congestion management, flexibility validation. Load and generation profiles of aggregated equivalents and distributed generators are frequently based on rough assumptions that systematically diverge from reality. The service provides the calibrated model foundation that all model-based grid analyses require.
Key Performance Indicators
MAE improvement at measured nodes vs. uncalibrated baseline
Physical plausibility of parameter estimates at unmeasured nodes
Validation accuracy on held-out period not used during calibration
Documentation clarity of calibrated vs. prior-knowledge-constrained nodes
Data Provided
Calibrated PowerFactory model with load/generation profiles reflecting real operating conditions
Parameter report: adjusted profiles, adjustment magnitude, data basis, direct vs. prior-knowledge calibration distinction
All calibration Python scripts fully documented and handed over
Inputs: PowerFactory model, PQM measurements at available measurement points, historical SCADA records
TEF
TEF DSO