What is unsupervised learning? Unsupervised Learning (see wikipedia) is used when no historical data is available that involves past faulty behaviour and/or ageing progression. Therefore,
Wear indicator of a transmission sub-system
A clutch’s component inside a gearbox tended to wear quicker than expected leading to dangerous situations. Analysing the device to check its wear condition required a full disassembly of the gearbox meaning high engineering costs and expensive downtime.
A vibration sensor was placed close to the clutch’s component. Customer asked for a wear indicator of the clutch’s component in order to be alerted when it is the right time to replace the component.
Diagfit was used to build, based on few healthy clutch’s component data, a prediction model providing a wear indicator.
Wear indicator led to anticipate, 25 flight hours ahead of time, a highly dangerous wear level. Thus allowing to raise an alert and schedule the work order to replace the clutch’s component.