End User
ELGO
Description
The ML-Based Outage Root Cause Identification service uses machine learning to identify patterns in large volumes of SCADA event logs and relate ongoing outage events to previously resolved incidents. By comparing current event sequences with historical incidents, the service returns the most similar past cases with their likely root causes, supporting faster interpretation of outage situations and more consistent disturbance assessment by distribution system operators.
Core Capabilities
Monitoring & Anomaly Detection
Predictive & Prescriptive Analytics
Business Need
In daily operation, SCADA-connected devices generate large volumes of event logs, while operators still rely on manual interpretation and field reports to understand outage causes. The process is time-intensive and prone to delays, particularly for complex multi-device events. The service helps operators analyse outage-related events more efficiently, relate them to historical incidents, and improve the speed and quality of disturbance assessment.
Key Performance Indicators
Top-k accuracy: proportion of cases where the correct outage cause appears in the top suggested results
Description similarity: closeness of suggested cause descriptions to documented incident descriptions
Categorisation consistency: ability to group outages into meaningful categories (equipment failure, vegetation, animal contact, human damage)
Low-confidence flagging rate: proportion of cases correctly identified as requiring manual review
Data Provided
Structured outage cause classifications with references to similar historical incidents
Short summaries of event patterns supporting proposed interpretation
Outage category assignments (equipment failure, vegetation, animal contact, human damage)
Confidence flags for cases requiring manual expert review
TEF
TEF TSO