Digital Twin System Construction for Steel Pipe Straightening Machine

Literature Overview

The paper by Dong Yunxiang, Chen Jianxun, Hu Jianhua, and Wang Chen, published in Shanxi Metallurgy, Volume 48, Issue 10, 2025, presents the development of a digital twin system for a six-roller steel pipe straightening machine used in the seamless steel pipe manufacturing process. The research is funded by the Shanxi Provincial Natural Science Foundation (Grant No. 202203021221158). The study addresses the transition from automation to intelligent manufacturing in heavy mechanical equipment, with a focus on the seamless steel pipe straightening process.

Technical Architecture and System Design

The digital twin system is constructed based on four dimensions: physical entity, virtual model, service interface, and twin data. The system framework is designed to enable real-time data exchange between the physical straightening machine and its virtual counterpart, achieving virtual-real synchronisation and data visualisation.

The following table summarises the key components of the digital twin system:

The PROFINET communication protocol is a key enabler of the digital twin system, as it provides high-speed, deterministic data exchange between the PLC and the digital twin platform. This ensures that the virtual model accurately reflects the real-time state of the physical machine, including roller positions, pipe dimensions, straightening parameters, and process variables.

Four-Dimensional Digital Twin Model

The four-dimensional model proposed in this paper is a structured approach to digital twin development:

1. Physical Entity Dimension: The actual six-roller straightening machine, including the rollers, drive system, sensors, and control hardware. This dimension captures the real-world machine state through sensor data.
2. Virtual Model Dimension: The 3D geometric and kinematic model of the straightening machine, constructed in Unity3D. This model represents the physical machine in a virtual environment and enables visualisation and simulation.
3. Service Interface Dimension: The communication interfaces and data exchange protocols that connect the physical and virtual dimensions. This includes the PROFINET network, PLC data points, and database interfaces.
4. Twin Data Dimension: The data layer that stores, processes, and analyses the data flowing between the physical and virtual dimensions. The MySQL database serves as the central data repository.

Engineering Practice and Process Implications

For the seamless steel pipe straightening process, the digital twin system offers several practical benefits:

• Process optimisation: By monitoring real-time straightening parameters, the system can identify optimal roller configurations for different pipe sizes and materials, reducing trial-and-error adjustments.
• Quality control: The virtual model can predict the straightness and dimensional accuracy of the pipe based on the current roller settings, enabling proactive quality assurance.
• Maintenance planning: The system can monitor the wear of rollers and other critical components, enabling predictive maintenance and reducing unplanned downtime.
• Operator training: The virtual environment can be used for operator training without affecting production, reducing the learning curve for new operators.

From a metallurgical perspective, the straightening process introduces residual stresses in the pipe material, which can affect subsequent welding operations and the overall mechanical properties of the finished pipe. The digital twin system can help optimise the straightening parameters to minimise residual stress levels, which is particularly important for high-strength steel grades and thick-walled pipes where residual stress can influence hydrogen-induced cracking susceptibility and fatigue performance.

Key Questions and Reflections

The paper presents a well-structured approach to digital twin development, but several questions arise regarding the practical implementation. First, the accuracy of the virtual model depends on the fidelity of the physical model, which requires detailed knowledge of the machine dynamics, material properties, and process parameters. The level of detail in the virtual model should be sufficient to capture the key process variables but not so detailed as to cause computational overhead.

Second, the integration of the digital twin system with existing manufacturing execution systems (MES) and enterprise resource planning (ERP) systems is not discussed in detail. In a real production environment, the digital twin must interoperate with other systems to provide end-to-end process optimisation.

Third, the paper mentions compatibility with intelligent algorithm approaches and neural network self-learning models, which suggests future extensions for process parameter optimisation and anomaly detection. However, the current implementation appears to focus on data visualisation and virtual-real synchronisation, which is a foundational step towards more advanced intelligent manufacturing capabilities.

Study Insights and Outlook

This paper demonstrates a practical approach to digital twin development for a specific manufacturing process. The use of established technologies such as PROFINET, PLC, MySQL, and Unity3D ensures that the solution is technically feasible and deployable in an industrial environment. The four-dimensional model provides a clear framework for digital twin development that can be adapted to other manufacturing processes. The integration of the digital twin with the straightening process offers tangible benefits in terms of quality control, process optimisation, and maintenance planning. Future research should focus on enhancing the virtual model fidelity, integrating advanced analytics for process optimisation, and extending the digital twin to cover the entire seamless steel pipe manufacturing chain, from hot rolling through to final inspection.