CASTeC™ is a software implementing data-driven diagnostics of companion elements such as, originally, the satellites of a constellation, through the early detection of anomalies and relevant events that require investigation by the elements controllers, and the automatic identification of correlated telemetries and events (including the discovery of new unknown correlations).
This data driven space born technology can be applied to any complex system composed of one of a kind or several companion subsystems of similar design.
Figure 1. CASTeC™ Core Modules exploitation in the test and operations phases
The increasingly high number of flying satellites makes it ever more difficult or impracticable for Satellite Controllers (SCs) to monitor effectively the health status of satellites, especially in large constellations.
A similar evolving difficulty applies to any complex system made of similar companion components such as those belonging to large hydrocarbon processing facilities, having similar specifications and/or designs.
CASTeC™ is a software tool improving monitoring and diagnostics, by:
- Early identification of anomalies in the behaviour of the system relative to a contextualised standard, characterised by specific operations or test conditions (e.g. in the space domain: operational mode, mission/production phase, test environment, manoeuvres or specific known fault modes).
- Identifying correlated events, thus enabling fault isolation and mitigation strategies definition.
- Identifying critical operative conditions that may affect the system’s service performance.
Figure 2. Evolution of CASTeC™ Index as a result of parameter behaviour
Figure 3. Example of four correlated parameters
CASTeC™ diagnostic capabilities are based on an innovative Machine Learning based approach, applied to contextualised periods of systems operations. This checking approach supersedes traditional threshold-based approaches as it allows anticipating the detection of an anomaly, even when parameters stay within nominal bounds. Compared to conventional Machine Learning, it allows understanding of the reasons for an anomaly detected in the telemetry.
Figure 4. CASTeC™ HMI
CASTeC™ provides early alerts on the status of the system elements, based on a detailed analysis of each single telemetry of the set.
CASTeC™ identifies relevant events from the degree of anomaly detected.
CASTeC™ supports event troubleshooting by guiding the identification of potential cause-effect relations among events.
No need for experts’ knowledge
CASTeC™ algorithms do not need experts’ knowledge on Artificial Intelligence to be configured, but only limited information and knowledge to define the relevant operative conditions of the companion elements.
Figure 5. CASTeC™ HMI
With CASTeC™ the user can:
- Check the behaviour of a subsystem/component against other companions of the same system;
- Monitor and understand the subsystem/component behaviour during operations through the visualisation of: raw parameters timeseries, derived features, CASTeC Index and events through different views and maps.
CASTeC™ can be easily interfaced with different plant control operations platforms.
CASTeC™ is delivered as a service with different options: on cloud or on customer’s premises.
The web-based Human Machine Interface implementation allows the installation of CASTeC™ on a shared (or external) machine and the investigation of the telemetry from any different location, through a web browser.
CASTeC™ can be used as a standalone tool or integrated with other end users’ tools, through specific APIs, according to their needs.
For further information:
CASTeC has been developed by S.A.T.E. Srl and Planetek Italia Srl in the frame of contracts with the European Space Agency. The view expressed herewith shall in no way be taken to reflect the official opinion of the European Space Agency.
Read the article online at: https://www.hydrocarbonengineering.com/special-reports/05102021/castec-from-space-systems-to-production-plants/
You might also like
ADNOC has announced that it has formally closed the acquisition of a 24.9% shareholding in OMV AG (OMV), a global energy and chemicals group.