Towards real-time control of surface treatments by using digital image correlation. – INDIC

The project INDIC aims to develop a methodology using the stereo digital image correlation method (S-DIC) to measure strains generated by surface processes. This method is non-contact, can be used in the laboratory or in industry, and can be technically integrated into the monitoring of manufacturing processes. Furthermore, recent advances in the optimization of DIC algorithms mean that this method can now be used in a manufacturing process optimization loop. Real-time strain knowledge could significantly enhance: process control, process modeling and understanding the effects on parts during in-service operation. However, several scientific challenges need to be overcome before the S-DIC approach can be used in this context: (i) most surface treatments could cause significant strains that are challenging to characterize with DIC. (ii) Obtaining marker stability during the application of external forces on the component such as media and rollers could be difficult. (iii) Process execution speeds could require the use of high-speed cameras, which would necessarily be less resolving than conventional cameras, generating a loss of spatial information. (iv) Real-time process correction is only achievable with the implementation of methodologies that limit computation times. The project therefore aims to remove these scientific barriers. To achieve this, a combination of experiments, process modeling and the use of artificial intelligence (AI) will be implemented. Tools will be created to facilitate the use of DIC during manufacturing processes and the interpretation of results. Obtained strain maps will be linked with fatigue properties and detection of surface heterogeneities can yield insightful data related to initiation mechanisms. Furthermore, the use of AI will make it possible to identify updated process parameters so that defects can be corrected in real time.