Vulnerability Assessments of Induction Machine-Based Multistage Rolling Mill System Under Sensor Integrity Attacks
Vulnerability Assessments of Induction Machine-Based Multistage Rolling Mill System Under Sensor Integrity Attacks
复制标题
DOI：
10.1109/tii.2024.3370240
复制
发表时间：
2024-06
影响因子：
12.3
通讯作者：
Kun Hu;Jin Ye;Wenzhan Song
中科院分区：
文献类型：
--
作者：
Kun Hu;Jin Ye;Wenzhan Song
本文针对制造系统中日益增长的网络攻击威胁，对多级轧钢机系统在各种传感器完整性攻击下的脆弱性进行了评估。我们首先给出了整个系统的详细建模。然后，设计了五种典型的完整性攻击，以模拟对厚度传感器、速度传感器和活套角度传感器可能的网络威胁。为了综合评价影响，我们提出了一个脆弱性评估框架，该框架包括1)5个设备级评估指标，用于评估制造质量、操作安全和铣削生产率；2)系统级指标，以反映对多级系统的综合影响。为了验证所提出的评估方法的有效性，在MATLAB/SIMULINK中对五级轧机系统中不同的传感器攻击场景进行了仿真。所提出的指标成功地评估了不同攻击案例所造成的影响，并显示了未来将所提出的指标应用于攻击检测和缓解的可能性。
In this article, we provide vulnerability assessments for a multistage rolling mill system under various sensor integrity attacks in response to the increasing cyber-attack threats in manufacturing systems. We first present detailed modeling of the whole system. Then, five typical integrity attacks are designed to simulate the possible cyber threats to the thickness sensor, speed sensor, and looper angle sensor. To comprehensively evaluate the impact, we propose a vulnerability assessment framework that includes 1) five device-level evaluation metrics to assess the manufacturing quality, operation safety and milling productivity, and 2) system-level indices to reflect the comprehensive impact on the multistage system. To verify the effectiveness of the proposed evaluation methods, simulations of different sensor attack scenarios in a five-stage rolling mill system are conducted in MATLAB/Simulink. The proposed metrics successfully assess the impact caused by different attack cases and show the possibility of applying the proposed indices for attack detection and mitigation in the future.