Predictive maintenance at the edge cuts bandwidth requirements
Dutch signal analysis firm Samotics is rolling out edge analytics for remote and offshore customers using its predictive maintenance software.
Performing analytics on the edge avoids the need for higher bandwidth links to process and transfer data, which makes it easily accessible in remote and offshore locations. The company recently signed a strategic deal with ABB to add its technology to drives and motors.
- Samotics raises €14.5m for industrial AI
- ABB takes stake in Samotics to expand AI predictive maintenance
- AI analytics tool boosts energy efficiency
The SAM4 Health software monitors the performance and efficiency of critical assets using electrical signature analysis (ESA). This analyzes the current and voltage signals of electric-driven equipment such as motors, pumps and conveyors from sensors in the motor control cabinet, rather than on the machine being monitored.
“The edge processing runs on the gateway devices that’s part of our hardware set-up. The device is used for sample selection, calculation of various failure indicators and data communication to the cloud,” said Thijs Bootsma, Head of Asset Health at Samotics.
“The gateway is installed in the motor control cabinet (MCC) as well as our other hardware devices. In some cases, the gateway is installed in a separate cabinet, when we monitor assets at the same location across several MCCs.”
This uses a similar approach on the edge as the cloud. Previously, all the raw data was shared to the cloud to compute the failure indicators that fuel the detection algorithms. Those calculations have been moved to the edge gateway, and the failure indicators are communicated to the cloud instead of the raw data. This reduces the data volume communicated to the cloud and failure indicators can be calculated more frequently for certain use cases, says Bootsma.
For many organizations with industrial assets located in remote geographies with limited and unreliable network connectivity, the amount of asset health data generated by connected devices can easily overwhelm their centralized data infrastructure. Edge analytics helps solve this challenge by collecting, storing, processing and analyzing data at the location where it is created, on the devices at the edge.
In addition to remote monitoring, remote and offshore customers can now achieve greater control over data movement and storage by creating, hosting, and processing data in one place. With data regulation increasing, this centralized approach can help organizations meet stringent data privacy and data sovereignty requirements.
“We’re always looking to improve and enhance our services and we’ve seen the potential of edge analytics to help our customers unlock the power of their data. Therefore, we’ve invested in enhancing our data processing capabilities on the edge to reduce our data communication while being able to generate failure indicators near real-time,” said Bootsma.
“This allows us to offer a complete solution to customers in remote, low-bandwidth locations. We look forward to selectively rolling out this feature in the coming months to our customers who experience network restrictions.”
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
