
ABSTRACT The escalating sophistication of cyber‐attacks and the widespread utilisation of stealth tactics have led to significant security threats globally. As digital twin systems are increasingly deployed on endpoints and edge nodes, they have become attractive targets for sophisticated malware attacks. However, existing endpoint detection and response (EDR) solutions often incur substantial runtime overhead, making real‐time deployment in resource‐constrained digital twin environments challenging. Nevertheless, the existing static detection methods exhibit limited coverage, and traditional dynamic monitoring approaches encounter challenges in bypassing evasion techniques. Thus, it has become imperative to implement nuanced and dynamic analysis to achieve precise behaviour detection in real time. There are two pressing concerns with current dynamic malware behaviour detection solutions. First, data collection and processing introduce significant overhead, limiting real‐time deployment on end hosts. Second, these approaches often treat malware as a single entity, overlooking varied behaviours within one instance. To address these gaps, we propose PARIS, an adaptive trace‐fetching, lightweight and real‐time malicious behaviour detection system. PARIS monitors malicious behaviour with Event Tracing for Windows (ETW) and selectively collects maliciousness‐related APIs or call stacks, significantly reducing collection overhead. As a result, it can monitor a wider range of APIs and detect more intricate attack behaviours. We implemented a prototype of PARIS and evaluated system overhead, comparative behaviour recognition accuracy and the impact of different models and parameters. Results show that PARIS reduces over 98.8% of data compared to raw ETW traces, thereby lowering memory, bandwidth and CPU overhead whilst maintaining a detection accuracy of 93.6%. Furthermore, breakdown evaluation shows that 80% of the memory and bandwidth savings and the complete CPU usage reduction come from our adaptive trace‐fetching collector. The lightweight design of PARIS makes it particularly suitable for protecting digital twin endpoints and edge nodes, where computational resources are limited and real‐time responsiveness is critical.
ABSTRACT Emergency evacuation in underground coal mines is hindered by the inability of static escape plans to adapt to rapidly changing post‐disaster conditions. This study constructs a digital twin‐oriented simulation framework integrating ultra‐wideband (UWB) personnel positioning with multi‐parameter environmental monitoring, and develops a dynamic evacuation path optimisation algorithm on a time‐varying hazard‐weighted graph. No physical mine is connected, so the framework is exercised entirely in simulation. Internal synchronisation latency remains within 2 s under disaster mode, and a simplified gas dispersion model deviates approximately 8.6% from literature benchmarks. A weighted A* algorithm variant incorporating a risk‐aversion heuristic and roadway capacity constraints is proposed. Under a gas outburst scenario with 260 personnel, the algorithm reduces mean evacuation time by 20.6% and cumulative hazard exposure by 34.2% compared to the static Dijkstra baseline, with a mean planning time of 1.4 s satisfying real‐time requirements. A coupled outburst and collapse case raises the number of hazardous segments from 7 to 11 through reduced dilution. Ablation experiments confirm the synergistic benefit of fusing positioning and environmental data, and sensitivity analyses verify the robustness of key weight parameters. Under combined sensing degradation, the two gains narrow to 18.8% and 31.1%, and they fall to 17.7% and 29.4% when route compliance drops to 70%. The findings provide a methodological reference for intelligent emergency evacuation in underground mines, though field validation with real mine data remains necessary.
ABSTRACT Physics‐informed neural networks (PINNs) provide a promising paradigm for integrating governing physical laws with data‐driven learning in digital twin systems. However, existing studies and reviews have primarily emphasised offline prediction accuracy or domain‐specific applications, whereas the structural role of PINNs in real‐time model updating and continuous synchronisation remains insufficiently clarified. This review focuses on real‐time updating mechanisms in evolving digital twins and examines how PINN‐based and operator‐learning‐based models support synchronisation between physical assets and their digital counterparts. First, the functional requirements of real‐time digital twin updating are analysed, including online calibration, data–model synchronisation, multitimescale response and application‐dependent latency constraints. Classical and advanced real‐time estimation methods are also compared to clarify their trade‐offs in nonlinearity handling, uncertainty quantification, computational cost and real‐time applicability. Second, the technical mechanisms of PINNs are reviewed from the perspectives of physics‐constrained parameter identification, time‐dependent learning, data–physics fusion, adaptive training and neural operator‐based generalisation. Third, a functional layered framework is proposed to organise PINN‐driven digital twins into interacting components responsible for physical consistency enforcement, synchronisation and data assimilation, online updating and computational acceleration. The methodological roles of PINNs are further clarified by distinguishing their functions as physics‐constrained surrogate models, inverse identification tools and data–physics fusion frameworks. Finally, structural challenges are critically synthesised, including optimisation instability, uncertainty propagation, sparse sensing and identifiability, synchronisation constraints, computational scalability and robustness under distribution shifts. Rather than treating PINNs as general‐purpose prediction models, this review positions them as structural mechanisms for regulating consistency, adaptability and computational efficiency in real‐time digital twin architectures.
ABSTRACT The utilisation of associated gas for power generation in offshore oilfields is essential for improving resource efficiency, yet operations remain hampered by substantial gas flaring and heavy reliance on manual experience. This paper presents a digital twin‐based predictive control strategy for multiplatform natural gas distribution, comprising four modules: (1) Digital twin module: Builds high‐fidelity digital models with a hybrid mechanistic‐data approach to simulate and monitor gas distribution systems; (2) operating condition prediction module: Uses a benchmark library from equipment models, combined with power demands and real‐time data, to predict valve openings, platform pressures and identify the control valve requiring adjustments; (3) intelligent distribution module: Integrates steady‐state models with self‐tuning PID algorithms to autonomously generate valve control schemes and (4) scheme verification module: Validates control schemes using dynamic models, providing feedback for parameter optimisation. A case study at a Chinese offshore oilfield demonstrates the strategy's effectiveness. The Lambda tuning method enabled PID self‐tuning, generating valve control schemes within 1 minute in response to target power loads, ensuring rapid and stable control. Gas flow deviations between dynamic simulation and setpoints remained below ± 5%, confirming practical applicability. This strategy offers a reliable solution for intelligent gas distribution as offshore operations advance towards unmanned platforms.
ABSTRACT Effective spatial data maintenance during the operation and maintenance phase of airports is imperative for the infrastructure safety, operational effectiveness, and compliance with regulations. However, airports encounter serious challenges such as lack of management support, insufficient funding, and poor‐quality data sources. Despite the recognised potential of emerging technologies such as digital twins, digital threads, ontologies, scalable data architectures, and governance frameworks, their integrated application for airport spatial data management remains unexplored. The existing literature focuses on analysing these technologies individually, leaving a large gap for their combined application for end‐to‐end comprehensive spatial data maintenance. To bridge this gap, this research proposes a comprehensive, literature‐based framework using qualitative research as the underlying approach with the inclusion of a comprehensive literature review, and semi‐structured interviews with experts for practical significance and theoretical richness. By employing inductive thematic analysis, we identified critical drivers and challenges in the practice of spatial data maintenance. The proposed framework brings together digital twins for realistic infrastructure simulation, digital threads for end‐to‐end data traceability, ontologies for semantic interoperability, data lakehouses for cost‐optimisation, and a governance layer for compliance and data integrity. The integrated ecosystem addresses the identified challenges by leveraging the identified drivers and technological synergies, enhancing data‐driven decision‐making capabilities across diversified airport settings. Although the proposed framework is theoretically validated, it must be tested on real airports, representing a critical limitation. Additionally, the scope of the research is constrained by stakeholder insights primarily from medium‐sized U.S. airports, which may limit applicability for large airports or other regulatory frameworks. Future research will involve piloting the framework for empirical testing of performance and adaptability, enhancing stakeholder engagement across the world, examining the strategies for mainstreaming with existing airport systems, and addressing ethical and cybersecurity issues.