• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2026, Vol. 62 ›› Issue (11): 48-60.doi: 10.3901/JME.260588

• 特邀专栏:制造互联与工业智能 • 上一篇    

扫码分享

数字孪生驱动的承压设备设计制造运维跨阶段数据集成方法

李孝斌1,2, 黄文明1,2, 江沛1,2, 尹超1,2, 曹华军1,2   

  1. 1. 重庆大学机械与运载工程学院 重庆 400044;
    2. 重庆大学高端装备机械传动全国重点实验室 重庆 400044
  • 收稿日期:2025-06-05 修回日期:2025-11-01 发布日期:2026-07-29
  • 作者简介:李孝斌(通信作者),男,1987年出生,副教授,博士研究生导师。主要研究方向为智能制造、网络协同制造,以及大数据、物联网、人工智能等新一代信息技术在离散制造行业中的应用。E-mail:xiaobin_lee@cqu.edu.cn;黄文明,男,2001年出生,硕士研究生。主要研究方向为智能制造。E-mail:1357788523@qq.com;江沛,男,1985年出生,副教授,博士研究生导师。主要研究方向为机器人技术、绿色制造、智能制造等。E-mail:peijiang@cqu.edu.cn;尹超,男,1974年出生,教授,博士研究生导师。主要研究方向为智能制造、网络协同制造、制造系统工程、新一代信息技术及应用。E-mail:ych925@cqu.edu.cn;曹华军,男,1978年出生,教授,博士研究生导师。主要研究方向为绿色制造、智能制造等。E-mail:hjcao@cqu.edu.cn
  • 基金资助:
    国家重点研发计划(2022YFB3306400)、国家自然科学基金(52475511,52075060)和重庆市技术创新重大研发(CSTB2024TIAD-STX0029)资助项目。

A Digital Twin-driven Method for Cross-stage Data Integration of Pressurized Equipment in Design, Manufacturing, and Operation & Maintenance

LI Xiaobing1,2, HUANG Wenming1,2, JIANG Pei1,2, YIN Chao1,2, Cao Huajun1,2   

  1. 1. College of Mechanical and Vehicle Engineering, Chongqing University, Chongqing 400044;
    2. State Key Laboratory of Mechanical Transmission for Advanced Equipment, Chongqing University, Chongqing 400044
  • Received:2025-06-05 Revised:2025-11-01 Published:2026-07-29

摘要: 数字孪生、人工智能等新一代信息技术与先进制造技术的深度融合应用,推动着量大面广的制造行业企业从传统生产型制造向服务型制造高质转型发展。然而,承压设备作为石化、冶金等流程工业生产的核心装备,其设计、制造与运维等环节普遍存在业务协作链条长、专业知识要求高、跨区域多业务间信息孤岛频现等问题,严重制约了全业务链条的高效协同运行与智能优化管控。为此,提出一种数字孪生驱动的承压设备全生命周期数据集成方法。首先,构建数字孪生驱动的承压设备全生命周期运行模式,阐明数字孪生驱动的全域信息共享机制;构建数字孪生驱动的承压设备设计-制造-运维跨阶段数据集成框架,研究全生命周期数据融合处理、模型虚实映射与语义对齐、人工智能驱动的数据价值挖掘等支撑该框架运行的关键技术;最后,通过某炼油厂管壳式换热器运维优化案例,验证所提方法的可行性与有效性。结果表明,该方法可有效提升承压设备跨阶段数据集成共享水平,提高设计、制造与运维效率,降低故障率与维护成本,为承压设备的智能化转型与可持续运维提供有力支撑。

关键词: 数字孪生, 承压设备, 全生命周期, 数据集成, 人工智能

Abstract: The integration of digital twin and artificial intelligence with advanced manufacturing technologies is driving the transformation of manufacturing enterprises from production-oriented to service-oriented models. However, pressure vessels, as core equipment in petrochemical and metallurgical industries, face significant challenges including lengthy collaboration chains, high technical requirements, and cross-regional information silos throughout their design, manufacturing, and operation phases. These issues severely limit efficient collaboration and intelligent control across the entire business chain. This paper proposes a digital twin-driven data integration method for pressure vessel full lifecycle management. A digital twin-driven operation mode is established with global information sharing mechanisms. A cross-stage data integration framework spanning design-manufacturing- operation phases is developed, incorporating key technologies such as lifecycle data fusion, virtual-real model mapping with semantic alignment, and AI-driven data mining. The method's feasibility and effectiveness are validated through a shell-and-tube heat exchanger optimization case in a refinery. Results show that the proposed approach significantly enhances cross-stage data integration, improves design and operational efficiency, reduces failure rates and maintenance costs, providing robust support for intelligent transformation and sustainable maintenance of pressure vessels.

Key words: digital twin, pressurized equipment, full lifecycle, data integration, artifical Intelligence

中图分类号: