A novel six-dimensional digital twin model for data management and its application in roll forming

Abstract

Roll forming is a cost-effective, efficient, and flexible plastic processing technique that involves the gradual transverse bending of metal strips using sequentially arranged forming dies. It is an important technological approach in response to lightweight, energy efficiency, and safety in various sectors, including new energy vehicles, aerospace, and rail transportation. However, the complexity of the process, diverse data types, and the effects of coupling and nonlinearity have led to challenges in process stability and quality control. The unclear impact mechanism of heterogeneous time-series data from multiple sources on product quality significantly hinders the development and widespread adoption of roll forming in industries. To facilitate the study of its mechanisms and optimize the control of forming quality, this paper introduces a digital twin (DT) model tailored for the roll forming field. It also presents a product-oriented feature data management framework based on the DT model. This framework facilitates feature-based data categorization across the complete lifecycle, enabling advanced data analysis in the roll forming domain. The feasibility and advantages of the proposed model and framework are validated through application to produce hat-shaped components. It is hoped that the work can provide valuable insights to the digitalization and intelligent transformation of roll forming field.