SmartVision™ is a Multi-fleet Remote Diagnostics & Predictive Condition-based Maintenance System for rolling stock and track. At is heart is Humaware’s Adaptive Anomaly Detector to provide automatic, reliable and early alerts with advanced algorithms and analytics to predict likely remaining useful life estimates.
Enhanced Track Circuit Monitoring
In collaboration with Network Rail, as part of their Predict and Prevent strategy, Humaware has delivered breakthrough data-driven diagnostics technology for Track Circuit Monitoring. The capability has been demonstrated to detect defects in track circuits on average 10 hours earlier than existing fixed threshold methods and has removed the requirement of using the five fixed thresholds which have to be setup and maintained individually for over 18,000 track circuits being monitored.
Health and Prognostics Assessment for Escalators
Telent, Humaware and the London Underground asset engineers worked jointly to develop the predictive analytics software using an open architecture. One immediate analysis revealed that a problematic escalator fault could have been detected five months before a major failure actually happened. It became obvious that a major benefit would be a re-scheduling and management of the task to maintain assets that are in use for 20 hours a day.
Constant False Alarm Rate (CFAR) Autotrend Evaluation Report
A collaborative program was successfully conducted by Humaware and the Aviation and Missile Research, Development, and Engineering Center (AMRDEC) in the US to assess the performance of the Constant False Alarm Rate (CFAR) Autotrend dynamic alert detection technology (now known as the Adaptive Anomaly Detector). The principal findings of the assessment were:
- Exceeds the fixed threshold detection accuracy.
- Provides a significantly earlier detection than the fixed thresholds.
- Detects maintenance events (material faults) that are not identified by fixed thresholds.
- Earth Leakage Detectors
- Points machines
- Wind Turbines
- Helicopter gearboxes