基于数字化班组现场作业场景的边云应用协同机制研究

Research on Edge Cloud Application Collaboration Mechanism Based on Digital Team On-Site Operation Scenarios

摘要: 数字化班组具有班组类型多、覆盖范围广、通信方式灵活多样、数据量大、数据模型复杂、缺乏公共的模型语义基础、模型架构可扩展性差等突出特点，传统中心化的物联网架构难以满足大规模异构场景下数字化班组现场作业的应用需求，因此需要面向云这一复杂多变的环境，其边云应用协同机制除了需要确保云端资源、服务被合法的用户所获取并使用之外，同时需要兼顾隐私保护、安全创建、可信自毁等问题。首先通过研究国内外边云协同机制的发展现状及典型案例，分析边云应用协同的总体能力与内涵，明确边云应用协同的参考框架，然后识别边云应用协同的模式和边云应用协同机制的实现方式，最后结合设备故障问题诊断、设备健康诊断与故障预测、负荷趋势预测、设备过热自动化判定等数字化班组现场作业场景的典型特征，设计了云–边协同、边–边协同、边–端协同、云–云协同的应用协同机制。

Abstract: The digital team has prominent characteristics such as multiple team types, wide coverage, flexible and diverse communication methods, large data volume, complex data models, lack of a common model semantic foundation, and poor scalability of model architecture. The traditional centralized IoT architecture is difficult to meet the application needs of digital team on-site operations in large-scale heterogeneous scenarios, so it needs to face the complex and ever-changing environment of the cloud, the edge cloud application collaboration mechanism not only needs to ensure that cloud resources and services are obtained and used by legitimate users, but also needs to con-sider issues such as privacy protection, security creation, and trusted self destruction. Firstly, by studying the current development status and typical cases of edge cloud collaboration mechanisms at home and abroad, the overall capability and connotation of edge cloud application collaboration are analyzed, and the reference framework for edge cloud application collaboration is clarified. Then, the patterns of edge cloud application collaboration and the implementation methods of edge cloud application collaboration mechanisms are identified. Finally, combined with equipment fault diagnosis, equipment health diagnosis and fault prediction, load trend prediction typical features of digital team on-site operation scenarios such as equipment overheating automation judgment have been designed with application collaboration mechanisms such as cloud edge collaboration, edge collaboration, edge end collaboration, and cloud collaboration.