Digital Twin Driven End-Face Defect Control Method for Hot-Rolled Coil With Cloud-Edge Collaboration

End-face defect control (EF-DC) of the hot-rolled coil is crucial to the quality management. The core of EF-DC is to identify defects accurately, predict defects in advance, and improve production in time to prevent similar defects. The current research works on EF-DC are mainly at the defect recognition stage, which merely rely on the operating data in physical space. However, different types of defects exist in coils with different reasons, therefore it is difficult to accurately control defects in time only through the hot rolling operating data. In order to solve the above problems, first an EF-DC framework based on digital twin with cloud-edge collaboration is designed in this article. The computing tasks are reasonably allocated through cloud-edge collaboration to realize the timeliness of control. Second, the virtual model of hot-rolled coil is constructed from four aspects: geometry, physics, behavior, and rules. Through the analysis of the defect mechanism, the abnormal behavior events are obtained, and the mapping relationship between the defect and the behavior event is established to realize the defect traceability and control. Finally, the feasibility of the proposed method is verified by taking the edge scratch defect as an example.

I. Introduction

As one of the indispensable raw materials for the manufacturing industry, the quality of hot-rolled coil is crucial for many types of end products, especially high-end products in aerospace, automotive, and precision instruments fields. In addition to surface and shape defect, end-face defect (including edge crack, thorn, damage, and scratch) is another important affecting factor, such as corrosion resistance, wear resistance, and fatigue strength [1]. Therefore, the end-face defect control (EF-DC) is getting more and more attention in Baowu, Nippon, Posco, Nucor, and other major steel enterprises. The core of EF-DC is to recognize defects, analyze causes, predict developing trend in advance, and take corresponding measures timely to prevent similar defects in subsequent production. EF-DC mainly includes the following three aspects:

Defect recognition: Identify whether there is a defect and its type. The current recognition method mainly focuses on surface or shape defect of hot-rolled coil, lacking standard end-face defect feature set, and related identification algorithm.

Defect analysis and prediction: Count the defects in the same batch (such as type, quantity, severity, etc.), analyze the cause of defects, and predict whether defect is controllable through data. It indicates that defects are most likely to occur in subsequent production if out of control.

Defect control: Formulate control methods and take corresponding measures to optimize and adjust production to prevent similar defects in subsequent production based on defect analysis and prediction.