In FADOMAS, industrial data is collected and analysed to discover the transitions between the different system states. Such interconnections are normally hidden under the streams of raw data. The vision of FADOMAS is multifold. First, the project investigates the conditions under which the determination of faults can be accommodated. This step validates state-of-the-art change detection schemes for the online fault diagnosis over various system parameters. As a second step, event correlation methods are examined to unveil the internal dynamics that govern the operation of the system (e.g., nodes failure, malfunctioning, under-performing, unexpected behaviour). As a final step, a root-cause analysis methodology is applied to (i) improve the situation awareness of the administrators, (ii) provide explanations about the current status, and (iii) recommend mitigation actions such as proactive maintenance that will minimise the downtime of the machinery.
FADOMAS is currently implementing the following standards for the connection with SCADA systems and sensors:
- IEEE C37.1-2007 – IEEE Standard for SCADA and Automation Systems. This standard specifies best practices design, implementation, and long-term management ofsupervisory control and data acquisition (SCADA) systems.
- OGC Sensor Observation Service (SOS), Observation and Measurements (O&M) and SensorML. These standards define a data model for the management of sensor data values coming from the machinery.