Automatic real-time fault detection for industrial assets using metasensors

Abstract

Large-scale industrial plants require physical sensors to continuously measure quantities such as temperatures or pressures. A large number of sensors is required to accurately describe the operating state of the plant, which unfortunately makes it very difficult for them to be effectively monitored by human operators. In this work we present a method to construct so-called metasensors, virtual sensors that compress the information from several sensors in an optimal manner. These metasensors are used as inputs to a novel anomaly detection system that automatically alerts operators to abnormal operation behaviour.

Perspectives on a Global Observing System to Assess Ocean Health

“…It is also possible to integrate the indicators of marine ecosystem state into a meta-indicator that provides a metric of the risk of failure to remain healthy of a particular ecosystem. This is an approach developed in industrial failure analysis, where multiple sensors are integrated into a meta-sensor, defined as a virtual sensor that compresses the information from several sensors in an optimal manner (Butters et al, 2015). Integrated, meta-indicators have been developed in the context of marine biodiversity assessments in European seas (e.g., Andersen et al, 2014;Borja et al, 2016;Heiskanen et al, 2016;Uusitalo et al, 2016).…”