Research Article
Cloud-Based Data-Driven Behavior Model Recovery for Distributed Automation Systems
@ARTICLE{10.4108/dtip.8591, author={Xian Wu and Chuanyang Yu and Likuan Zhang and Hui Zhang and Wenbin Dai}, title={Cloud-Based Data-Driven Behavior Model Recovery for Distributed Automation Systems}, journal={EAI Endorsed Transactions on Digital Transformation of Industrial Processes}, volume={1}, number={1}, publisher={EAI}, journal_a={DTIP}, year={2025}, month={3}, keywords={Model-Driven Engineering, IEC 61499 Function Blocks, Model Recovery, Finite-State Machine, Data-Driven Model Generation}, doi={10.4108/dtip.8591} }- Xian Wu
Chuanyang Yu
Likuan Zhang
Hui Zhang
Wenbin Dai
Year: 2025
Cloud-Based Data-Driven Behavior Model Recovery for Distributed Automation Systems
DTIP
EAI
DOI: 10.4108/dtip.8591
Abstract
Industrial cyber-physical systems provide a bridge between legacy controllers and new edge devices that are usually equipped with massive computing power and storage capacity. The migration from legacy control systems to industrial cyber-physical systems is facing challenges as control code and design documents of legacy systems may not be available. This paper proposes a data-driven behavior model recovery method for the black-box distributed manufacturing system based on cloud computing. This method adopts the IEC 61499 function blocks as meta-models to describe system behaviors. The proposed framework includes three parts: data mining, logic restoration, and application construction. Raw collected data are processed and encapsulated into function block sets, then execution control charts, and finally function block types. This model recovery method is validated with a process control system of the food and beverage industry. A deployable function block network is generated by instantiating and connecting these function blocks.
Copyright © 2025 X. Wu et al., licensed to EAI. This is an open access article distributed under the terms of the CC BY-NC-SA 4.0, which permits copying, redistributing, remixing, transformation, and building upon the material in any medium so long as the original work is properly cited.
