Our client has been making sustainable progress possible and driving positive change on every continent. Customers turn to them to help them develop infrastructure, energy, and natural resource assets.
Client Industry Background
Problems
- Check alignment manually is time-consuming and labor-intensive.
- Errors in manual alignments.
Problem Implications
- If the gears are not aligned properly, the load may not be distributed evenly which eventually, cause gear teeth damage
- Gear slip may occur, which will result in equipment inefficiency
Client Requirements
- To automate the inspection of alignment of the gear with the help of machine vision to achieve high accuracy. Meshed gears are a part of a certain machine.
- To reduce the inspection cycle time, from 9-12 seconds to 1 second
Current Process
Manual Inspection is being carried out with multiple operators in various assembly lines.
Solution Using Machine Vision And AI
A camera or set of cameras with a specialized lens and appropriate illumination (red lights in this case) is set up to identify whether the gears are aligned or not. Images are captured and sent to the software (Qualitas EagleEye® Platform) cloud where the training is done using the Deep Learning algorithms. Once the program is trained, a real-time alignment check takes place, based on which the results are sent to PLC to take action.
Here the Anomaly detection tool is used to identify the defects.
Setup
IMAGES
CHECKING ALIGNMENT IN QEP(QUALITAS EAGLE-EYE® PLATFORM)
Conclusion
POC(Proof Of Concept) is conducted and the following conclusion is observed:
- Alignment inspection time is reduced to 0.75 second
- Operators are reduced in the inspection station
- The accuracy of detecting aligned and misaligned gears is ~99%
