End User
SWW
Description
The AI-Phase Identification service determines to which of the three electrical phases (A, B, or C) each smart meter or consumer connection is physically attached in LV distribution networks. By analysing correlation patterns in smart meter measurements, the service delivers phase assignments, confidence scores, and aggregated phase-load statistics for transformers and feeders.
Core Capabilities
Monitoring & Anomaly Detection
Predictive & Prescriptive Analytics
Business Need
DSOs often lack reliable phase connectivity records because infrastructure upgrades and undocumented maintenance leave phase labels incomplete or outdated. Missing phase information hampers network monitoring, planning, load-balancing, and phase-imbalance detection — all increasingly critical as EV charging and heat pump adoption create significant single-phase load imbalances in LV networks.
Key Performance Indicators
Phase Identification Accuracy: proportion of inferred phases matching validated connectivity information
Phase Assignment Stability: consistency of assignments across consecutive analytical periods
Phase Load Balance Indicator: evenness of load distribution across the three phases per transformer
Data Provided
Per-meter records: inferred phase (A/B/C), confidence score, correlation score, load balance contribution, data quality flag
Transformer/feeder aggregated phase-load distribution statistics
Inputs: smart meter time-series (active/reactive energy), transformer-level phase measurements, asset metadata
TEF
TEF DSO