Abstract
Anomalies and faults can be detected, and their causes verified, using both data-driven and knowledge-driven techniques. Data-driven techniques can adapt their internal functioning based on the raw input data but fail to explain the manifestation of any detection. Knowledge-driven techniques inherently deliver the cause of the faults that were detected but require too much human effort to set up. In this paper, we introduce FLAGS, the Fused-AI interpretabLe Anomaly Generation System, and combine both techniques in one methodology to overcome their limitations and optimize them based on limited user feedback. Semantic knowledge is incorporated in a machine learning technique to enhance expressivity. At the same time, feedback about the faults and anomalies that occurred is provided as input to increase adaptiveness using semantic rule mining methods. This new methodology is evaluated on a predictive maintenance case for trains. We show that our method reduces their downtime and provides more insight into frequently occurring problems.
Highlights
Sensor monitoring systems are transforming the industry, steered by the so-called Internet of Things (IoT) and Artificial Intelligence (AI) fields, through game-changing applications in, e.g., transportation [1], security [2], ventilation [3] and healthcare [4]
We introduce FLAGS, the Fused-AI interpretabLe Anomaly Generation System, and combine both techniques in one methodology to overcome their limitations and optimize them based on limited user feedback
Two anomalies were relabeled as switches as described above, which led the third one to be automatically labeled as a switch event
Summary
Sensor monitoring systems are transforming the industry, steered by the so-called Internet of Things (IoT) and Artificial Intelligence (AI) fields, through game-changing applications in, e.g., transportation [1], security [2], ventilation [3] and healthcare [4]. A wide variety of sensors are used, such as accelerometers to monitor vibrations, ultrasonic, inductive and draw-wire range sensors, shock pulse sensors or gyroscopes for rotational speed. Monitoring these sensors can deliver valuable insights into the physical assets, the performance, and the interaction with the environment. Sensor monitoring systems analyze so-called sensor networks and can be used to detect faulty or deviating system behavior using methodologies such as Anomaly Detection (AD), Fault Recognition (FR) and Root Cause Analysis (RCA)
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call