Validation of Digital Twins to predict hot rolled strip profile with plant data

Purpose and goal

** Denna text är maskinöversatt ** The project aims to validate four digital twins with manually measured plant data to facilitate the production of strips with limited profile variation. The goal is to validate the Crown4 model, its extended AI model and the data analysis platform aCurve for the aluminum hot rolling process in a production environment.

Expected effects and result

** Denna text är maskinöversatt ** The digital twins validated in this project will help industrial plants achieve higher quality products and higher yields. This will not only improve the competitiveness of rolled aluminum products from Sweden, but also reduce the environmental impact of their products and reduce their demand for Sweden´s limited green electricity.

Planned approach and implementation

** Denna text är maskinöversatt ** This project validates and improves the accuracy of the model through optimization with help of measurements. First, plant data will be collected through carefully designed trials for specific alloys. Next, the model´s parameters will be fine-tuned to improve its accuracy, followed by validation with reserved measurements and plant data. Finally, the model and its user interface will be tested by end-users to validate its function as a sensitivity analysis tool and aid in the production of knitting schedules.