Modelling powder paint coverage for automated industrial painting applications

Canada is suffering from a labour shortage, and especially so for hands-on operational workers. Among such industrially-relevant work, powder coating needs extensive manual intervention due to the inherent geometrical variability of painted parts and the complexity of the painting process itself. It is becoming increasingly difficult to find qualified workers to powder coat parts, driving the need for an automated technological solution.

This project is part of a broader program to automate the powder coating of parts on a production line of high variety and low volume (e.g., personalized parts) in the context of industry 4.0. The objective of this project is to develop a physical model that will be used to train a neural network to accurately and rapidly paint miscellaneous parts via powder coating. The model must ensure full coverage of the part and uniform paint thickness. A review and evaluation of relevant simulation literature and methods will be required. The optimal method will be programmed, and its performance evaluated and improved through testing the model on a robot to powder coat real parts, first on simple geometries and later on more complex geometries.
 

Required knowledge

Good knowledge of programming. Knowledge of image processing a plus

Desired program of studies

Masters with project, Masters with thesis

Research domains

Sustainable Development, the Circular Economy and Environmental Issues, Innovative Materials and Advanced Manufacturing

Financing

Attractive industrial bursary (details on request)

Additional information

Starting: September 1,  2022

Partners involved : Cadence Automation Inc., Technologies NeurobotIA Inc.  

Other professor contact : Giuseppe Di Labbio

Contact person

Lucas Hof | lucas.hof@etsmtl.ca