Hierarchical Federated Transfer learning and Digital Twin Enhanced Secure Cooperative Smart Farming DOI

Lopamudra Praharaj,

Maanak Gupta, Deepti Gupta

et al.

2021 IEEE International Conference on Big Data (Big Data), Journal Year: 2023, Volume and Issue: unknown, P. 3304 - 3313

Published: Dec. 15, 2023

The agriculture industry is extensive utilizing AI and data-driven systems for efficiency automation, with the goal to meet rising food demand. Individual farm owners can leverage agricultural cooperatives consolidate resources, exchange data, share domain knowledge. These enable generation of AI-supported insights their member farmers. However, this collaborative approach has raised concerns among individual smart regarding cybersecurity threats, privacy. A breach not only endangers attacked but also risks entire network farms members within cooperative. In research, we emphasize security challenges cooperative farming introduce a multi-layered architecture incorporating Digital Twins (DT). Further, hierarchical federated transfer learning framework designed address mitigate threats in farming. Our leverages Federated Learning (FL) based Anomaly Detection (AD), which operate on edge servers, enabling execution AD models locally without exposing farm's data. This localization excellent generalization ability, highly improve detection unknown cyber attacks. We employ FL structure that supports aggregation at various levels, fostering multi-party collaboration. Furthermore, have devised an integrates Convolutional Neural Networks (CNN) Long Short-Term Memory (LSTM) models, complemented by learning. objective expedite training duration while upholding high accuracy levels. To illustrate our proposed architecture, present use case demonstrate model's capabilities. proof-of-concept implementation Amazon Web Services (AWS) environment, reflecting real-world feasibility.

Language: Английский

OpenTwins: An open-source framework for the development of next-gen compositional digital twins DOI Creative Commons
Julia Robles, Cristian Martín, Manuel Dı́az

et al.

Computers in Industry, Journal Year: 2023, Volume and Issue: 152, P. 104007 - 104007

Published: Aug. 22, 2023

Although digital twins have recently emerged as a clear alternative for reliable asset representations, most of the solutions and tools available development are tailored to specific environments. Furthermore, achieving complex often requires orchestration technologies paradigms such machine learning, Internet Things, 3D visualization, which rarely seamlessly aligned in open-source solutions. In this paper, we present an framework compositional twins, i.e., advanced that link individual entities or subsystems create higher degree twin, allowing knowledge sharing data relationships. open framework, can be easily developed orchestrated with 3D-connected visualizations, IoT streams, real-time machine-learning predictions. To demonstrate feasibility use case Petrochemical Industry 4.0 has been developed.

Language: Английский

Citations

34
Digital Twin-based manufacturing system: a survey based on a novel reference model DOI
Shimin Liu, Pai Zheng, Jinsong Bao

et al.

Journal of Intelligent Manufacturing, Journal Year: 2023, Volume and Issue: 35(6), P. 2517 - 2546

Published: July 18, 2023

Language: Английский

Citations

24
Anomaly diagnosis of connected autonomous vehicles: A survey DOI
Yukun Fang, Haigen Min, Xia Wu

et al.

Information Fusion, Journal Year: 2024, Volume and Issue: 105, P. 102223 - 102223

Published: Jan. 3, 2024

Language: Английский

Citations

12
Digital Twins for Anomaly Detection in the Industrial Internet of Things: Conceptual Architecture and Proof-of-Concept DOI Creative Commons
Alessandra De Benedictis, Francesco Flammini, Nicola Mazzocca

et al.

IEEE Transactions on Industrial Informatics, Journal Year: 2023, Volume and Issue: 19(12), P. 11553 - 11563

Published: Feb. 22, 2023

Modern cyber-physical systems based on the Industrial Internet of Things (IIoT) can be highly distributed and heterogeneous, that increases risk failures due to misbehavior interconnected components, or other interaction anomalies. In this paper, we introduce a conceptual architecture for IIoT anomaly detection paradigms Digital Twins (DT) Autonomic Computing (AC), test it through proof-of-concept industrial relevance. The is derived from current state-of-the-art in DT research leverages MAPE-K feedback loop AC order monitor, analyze, plan, execute appropriate reconfiguration mitigation strategies detected deviation prescriptive behavior stored as shared knowledge. We demonstrate approach discuss results by using reference operational scenario adequate complexity criticality within European Railway Traffic Management System.

Language: Английский

Citations

20
A review of digital twins and their application in cybersecurity based on artificial intelligence DOI Creative Commons
MohammadHossein Homaei, Óscar Mogollón-Gutiérrez, José Carlos Sancho Núñez

et al.

Artificial Intelligence Review, Journal Year: 2024, Volume and Issue: 57(8)

Published: July 10, 2024

Abstract The potential of digital twin technology is yet to be fully realised due its diversity and untapped potential. Digital twins enable systems’ analysis, design, optimisation, evolution performed digitally or in conjunction with a cyber-physical approach improve speed, accuracy, efficiency over traditional engineering methods. Industry 4.0, factories the future, continue benefit from provide enhanced within existing systems. Due lack information security standards associated transition cyber digitisation, cybercriminals have been able take advantage situation. Access product service equivalent threatening entire collection. There robust interaction between artificial intelligence tools, which leads strong these technologies, so it can used cybersecurity platforms based on their integration technologies. This study aims investigate role providing for versions various industries, as well risks versions. In addition, this research serves road map researchers others interested security.

Language: Английский

Citations

8
Anomaly Detection in Industrial Machinery Using IoT Devices and Machine Learning: A Systematic Mapping DOI Creative Commons
Sérgio F. Chevtchenko, Élisson da Silva Rocha,

Monalisa Cristina Moura Dos Santos

et al.

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 128288 - 128305

Published: Jan. 1, 2023

Anomaly detection is critical in the smart industry for preventing equipment failure, reducing downtime, and improving safety. Internet of Things (IoT) has enabled collection large volumes data from industrial machinery, providing a rich source information Detection (AD). However, volume complexity generated by ecosystems make it difficult humans to detect anomalies manually. Machine learning (ML) algorithms can automate anomaly machinery analyzing data. Besides, each technique specific strengths weaknesses based on nature its corresponding systems. portion existing systematic mapping studies AD primarily focus addressing network cybersecurity-related problems, with limited attention given sector. Additionally, related literature do not cover challenges involved using ML within context IoT ecosystems. Therefore, this paper presents study devices address gap. Our primary objective investigate use models an setting, particularly The comprehensively evaluates 84 relevant spanning 2016 2023, extensive review research. findings identify most commonly used algorithms, preprocessing techniques, sensor types. identifies application areas points future research opportunities.

Language: Английский

Citations

12
Digital twin-assisted intelligent anomaly detection system for Internet of Things DOI

Burcu Bolat-Akça,

Elif Bozkaya

Ad Hoc Networks, Journal Year: 2024, Volume and Issue: 158, P. 103484 - 103484

Published: March 26, 2024

Language: Английский

Citations

4
A Framework for Product Life Cycle Management Based Digital Twin Implementation in the Aerospace Industry DOI Creative Commons

Busra Oksuz Gurdal,

Özlem Müge Testik

Applied Stochastic Models in Business and Industry, Journal Year: 2025, Volume and Issue: 41(2)

Published: Feb. 19, 2025

ABSTRACT As an emerging technology, digital twin (DT) studies are gaining momentum in both academia and industry. Specifically, the aerospace industry can benefit significantly from implementation of DT technology since its products processes complex, technically challenging, costly. DTs enable a comprehensive integration capacity holistic approach product life cycle. However, for simplification, implementations to often handled independently without with other related processes. In this study, we propose methodological framework integrate different throughout essential parts aircraft's pursuit creating system managing cycle aircraft, all aspects have been thoroughly examined. Ten main components management identified. Statistical stochastic approaches enhancing analytical capabilities discussed. Within scope Product Life Cycle Management perspective Systems Engineering, advocate aircraft by combining each component through thread.

Language: Английский

Citations

0
Research on engine cooling system condition monitoring based on deep digital twin DOI Creative Commons
Jian Wang, Ying Huang, Liguo Wei

et al.

Proceedings of the Institution of Mechanical Engineers Part D Journal of Automobile Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 27, 2025

Cooling system is a crucial subsystem essential for the engines, and its condition monitoring plays an important role in engine safety reliability. This paper proposes innovative deep digital twin (DDT) model that combines Gradient Boosting Decision Tree (GBDT) based ensemble learning Stacked Sparse Autoencoder (SSAE) to enhance sensitivity accuracy of cooling (CM). The algorithm employed generate coolant temperature baselines healthy state under varying operational conditions. Then, taking as characteristic parameter, health feature extraction constructed using network extract representations from different states. Specifically, efficiency extraction, this introduced modifications structure SSAE. To assess quantitively, probability density function (PDF) was calculated, with Kullback-Leibler (KL) divergence serving indicators (HI). severity system’s degradation indicated by comparing deviation KL between Simulation experimental data validation demonstrate capability proposed method monitoring.

Language: Английский

Citations

0
The Future of Digital Twin Research and Development DOI
Douglas L. Van Bossuyt, Douglas Allaire, Jason Bickford

et al.

Journal of Computing and Information Science in Engineering, Journal Year: 2025, Volume and Issue: 25(8)

Published: April 16, 2025

Abstract While digital twin (DT) has made significant strides in recent years, much work remains to be done the research community and industry fully realize benefits of DT. A group 25 professionals, US federal government researchers, academics came together from 11 different institutions organizations identify 14 key thrusts 3 cross-cutting areas for further DT development (R&D). This article presents our vision future R&D, provides historical context DT’s birth growth as a field, examples DTs use lab, discusses current state research. We hope that this serves nucleation point R&D efforts with shared trajectory collectively advance so society can more rapidly see

Language: Английский

Citations

0