GPAINNOVA Applies AI To Metal Surface Finishing In An EU-Funded Project

GPAINNOVA, a technology group specializing in metal surface finishing, has reached the final phase of the AI4Electropolishing project, funded by the European Union through the Next Generation program.

The aim of this initiative is to leverage artificial intelligence (AI) to develop digital tools for detecting anomalies in dry electropolishing processes. Utilizing virtual sensors and digital twins (digital representations of objects or procedures), the system can automatically identify and address potential issues. Through the algorithms embedded in the configurations, users of the patented DryLyte Technology, developed and marketed by GPAINNOVA, will gain the capability to optimize, control, and automate their processes, thereby achieving optimal results.

Four simultaneous actions
This initiative, which began in late 2022 and will end in May 2024, revolves around the DLyte PRO500 surface finishing machine and pursues four objectives: 

1. Development of a virtual conductivity sensor. GPAINNOVA is developing an algorithm connected to data science, which, relying on the variables of each machine, enables the inference of the conductivity value of the consumable in each surface finishing process. This enhancement aims to streamline existing procedures. Xavier Almendros, project manager and data scientist at GPAINNOVA, and one of the project leaders, explains, “Previously, determining the electrolyte’s conductivity required stopping the machine and measuring parameters to assess the need for corrective actions. 
2. Creating control algorithms based on a virtual sensor of the quality of finishing. The researchers aim to precisely determine the roughness, material loss, surface microhardness, and corrosion level of each workpiece after undergoing treatment with DryLyte finishing technology. “Several algorithms have been proposed for this task, with the goal of evaluating the most suitable data architectures to achieve these values,” says Almendros. 
3. Creating a configuration recommendation algorithm. GPAINNOVA is working on obtaining a recommendation algorithm based on historical data, to propose power source configuration values and process time, considering the material, the electrolyte and the geometry of the part to be treated. In this way, it will be possible to improve the quality of electropolishing. 
4. Evalution and validation of the surface integrity of the DryLyte Technology on cobalt-chromium samples. 

The AI4Electropolishing project is promoted by Red.es, the entity under the Spanish Secretary of State for Digitalization and Artificial Intelligence, and has a budget of €540,298, with a non-refundable subsidy amount of €405,223.50.