Abstract:The rise of digital twin-based operational improvements poses a challenge to protecting industrial cyber-physical systems. It is crucial to safeguard digital twins while disclosing internals, which can create an increased attack surface. However, leveraging digital twins to simulate attacks on physical infrastructure becomes essential for enhancing ICPS cybersecurity resilience. This paper introduces an integrated intelligent defense framework called CyberDefender to study various attacks on digital twin-based ICPS from a four-layer perspective (i.e., digital twin-based industrial cyber-physical systems infrastructure layer, honeynet and software-defined industrial network layer, intelligent security platform layer, and smart industrial application layer). To demonstrate its feasibility, we implemented a proof-of-concept (PoC) solution using open-source tools, including AWS for cloud infrastructure, T-Pot for Honeynet, Mininet for SDN support, ELK tools for data management, and Docker for containerization. This framework utilizes an integrated intelligent approach to enhance intrusion detection and classification capabilities for digital twin-based industrial cyber-physical systems (DT-ICPS). The proposed intrusion detection system (IDS) combines two strategies to improve security. First, we present an innovative approach to identifying essential features using explainable AI and ensemble-based filter feature selection (XAI-EFFS). By using Shapley Additive Explanations (SHAP), we analyze the impact of different variables on predictive outcomes. Secondly, we propose a hybrid GRU-LSTM deep-learning model for detecting and classifying intrusions. We optimize the hyperparameters of the GRU-LSTM model by using a Bayesian optimization algorithm. The proposed method demonstrates excellent performance, outperforming conventional state-of-the-art techniques with an accuracy rate of 98.96%, which is a remarkable improvement. Additionally, it effectively detects zero-day attacks, contributing to digital twin-based ICPS cybersecurity resilience.

Reconfigurable Digital Twin to Support Research, Education, and Training in the Defense of Critical Infrastructure

A Framework for Digital Twin of Civil Infrastructure— Challenges & Opportunities

CyberDefender: an integrated intelligent defense framework for digital-twin-based industrial cyber-physical systems

A Digital Twin Framework for Cyber Security in Cyber-Physical Systems

Digital Twin in Safety-Critical Robotics Applications: Opportunities and Challenges

Digital Twins and the Future of their Use Enabling Shift Left and Shift Right Cybersecurity Operations

Digital Twin for Evaluating Detective Countermeasures in Smart Grid Cybersecurity

Digital Twins Serving Cybersecurity: More Than a Model: Cybersecurity as a Future Benefit of Digital Twins 2

Digital Twin: Values, Challenges and Enablers

Security-Enhancing Digital Twins: Characteristics, Indicators, and Future Perspectives

Digital Twin-based Intrusion Detection for Industrial Control Systems

Digital Twin of Industrial Robots Over the Cloud: Framework, Design and Analysis

A Digital Twinning Platform for Integrated Sensing, Communications and Robotics

Digital Twin Virtualization with Machine Learning for IoT and Beyond 5G Networks: Research Directions for Security and Optimal Control

Digital twins based on augmented reality of measurement instruments for the implementation of a cyber-physical system

Digital twin framework for reconfigurable manufacturing systems (RMSs): design and simulation

A Security Framework in Digital Twins for Cloud-based Industrial Control Systems: Intrusion Detection and Mitigation

EPICTWIN: An Electric Power Digital Twin for Cyber Security Testing, Research and Education

A Modular Digital Twinning Framework for Safety Assurance of Collaborative Robotics

Shaping the future of industry: Understanding the dynamics of industrial digital twins

A Survey on Digital Twin for Industrial Internet of Things: Applications, Technologies and Tools