Projects → Predictive Maintenance
In the industrial world, equipment outage is to be avoided at all costs. But how do you know if equipment is about to break? Lots of industrial equipment: pumps, actuators, fans, HVAC systems will start showing early signs of troubles by vibrating more. Via the use of accelerometer-based sensors, the vibration profile can be captured by an on-premise gateway and analyzed in the Azure Cloud, compared to the profile of a normally operating equipment and trigger a maintenance alert, if needed.
The solution is meant to work on a very simple principle: in the industrial world, lots of industrial equipment like pumps, actuators, fans and HVAC systems will start showing early signs of troubles by increasing the frequency or strength of vibrations. Now, equipment outage has to be avoided, so this system uses accelerometer-based sensors to create a vibration profile of the system, which can be captured by an on-premise gateway and analyzed in the Azure Cloud. Once the working profile is compiled, it can be used to detect malfunctioning equipment and trigger a maintenance alert, if needed. For this project, DataArt connected a sensor and Raspberry Pi to a fan to identify variances.
Predicting when industrial equipment is going to fail and solving the problem before it happens is one of the key promises for the industrial Internet of Things [IoT]. What if instead of slides, anybody can deploy an open source predictive maintenance solution in minutes on Microsoft Azure. DataArt is a leader in cloud, big data & IoT consulting and open sourced DeviceHive, the first IoT & big data platform that makes different IoT standards talk to any app. DataArt partnered with Canonical, the company behind Ubuntu, as well as Microsoft. DeviceHive integrates with Snappy Ubuntu Core and Juju Charms. Via the Azure Marketplace anybody can run the DeviceHive charms on Azure in minutes and as such is able to connect smart devices to a big data back-end almost instantly.