Scholarlink

<p>The Programmable Logic Controller (PLC) is a digital computer essential to industrial automation systems, managing a wide range of input and output devices. Despite its important role in industrial applications, maintaining and expanding the functionality of PLC programs continues to incur significant costs and effort. To address this challenge, deep learning-based models have been proposed for PLC restoration. While these models offer automation and accuracy, their black-box nature and dependency on data quality lead to issues with interpretability and reliability. In this work, we introduce a symbolic-based model, also called White Box Networks (WBN), to recover the functional blocks of PLC circuits. WBN leverages mathematical symbols and logical expressions to provide a clearer understanding of system structures, resulting in more interpretable and reliable models. By effectively reconstructing the internal structure of PLC programs, we expect it to reduce maintenance costs, enhance long-term reliability, and improve the efficiency of industrial automation systems, contributing to the continued advancement of critical automation devices.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Reverse engineering for programmable logic controller structure estimation via white box networks

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年8月1日

查看全文
  • Eun Saem Lee,
  • Hyeontae Jo,
  • Hyung Ju Hwang

摘要

The Programmable Logic Controller (PLC) is a digital computer essential to industrial automation systems, managing a wide range of input and output devices. Despite its important role in industrial applications, maintaining and expanding the functionality of PLC programs continues to incur significant costs and effort. To address this challenge, deep learning-based models have been proposed for PLC restoration. While these models offer automation and accuracy, their black-box nature and dependency on data quality lead to issues with interpretability and reliability. In this work, we introduce a symbolic-based model, also called White Box Networks (WBN), to recover the functional blocks of PLC circuits. WBN leverages mathematical symbols and logical expressions to provide a clearer understanding of system structures, resulting in more interpretable and reliable models. By effectively reconstructing the internal structure of PLC programs, we expect it to reduce maintenance costs, enhance long-term reliability, and improve the efficiency of industrial automation systems, contributing to the continued advancement of critical automation devices.