Digital Twin for Aircraft Wing Maintenance with ML-Based Deviation Detection

Abstract

A growing need in the aviation industry is the development of smart maintenance systems that have the ability to not only improve aircraft safety but also reduce operating costs and time. Existing inspection technologies are mostly time-based and do not allow one to find structural defects between maintenance intervals. Digital Twin (DT) technology allows obtaining a virtual copy of the asset in real-time through integration of sensors, computational models, cloud communication, and predictive analysis. In this work, we propose a DT-based framework for aircraft wing maintenance with ML-based deviation detection. Our framework includes a physical aircraft wing model with a Flex sensor and MPU6050 accelerometer connected to an ESP32 microcontroller for real-time data collection. Data from the sensors are transmitted to the ThingSpeak cloud service and synchronised with a virtual twin of the physical wing created using MATLAB/Simulink and Simscape. Decision Tree classifier uses strain and vibration characteristics to classify the state of the wing as normal or abnormal to perform predictive maintenance.

Introduction

The text examines the use of Digital Twin technology, IoT sensors, and machine learning for real-time structural health monitoring of aircraft wings. The main purpose is to improve aircraft safety and move maintenance from traditional scheduled inspections toward intelligent condition-based and predictive maintenance.

Background

Aircraft wings experience continuous aerodynamic loads, vibrations, cyclic stresses, and environmental effects, which can gradually cause cracks, deformation, fatigue, or other structural problems. Traditional maintenance relies heavily on inspections after a fixed number of flight hours. Although this approach supports regulatory requirements, it may:

• Detect structural problems later than desirable.

• Require unnecessary inspections or maintenance.

• Provide limited information about the actual real-time condition of the aircraft.

The study proposes using a Digital Twin to overcome these limitations.

Digital Twin approach

A Digital Twin is a digital representation of a physical system that is continuously updated using real-time sensor data. Unlike conventional simulations, it creates an ongoing connection between the physical aircraft wing and its virtual model.

The proposed system combines:

• IoT sensors

• ESP32 microcontroller

• Flex sensor for measuring wing bending/deformation

• MPU6050 accelerometer for measuring vibration

• ThingSpeak cloud platform for data communication and synchronization

• MATLAB/Simulink and Simscape for modelling the virtual aircraft wing

• Decision Tree machine learning for detecting abnormal structural behaviour

The physical wing sends live sensor measurements to the cloud, where the data are synchronized with the virtual wing model. The machine-learning system then determines whether the wing is operating normally or showing abnormal behaviour.

Research gap

Although Digital Twins are already being researched for aircraft manufacturing, structural monitoring, predictive maintenance, flight monitoring, and lifecycle management, the text identifies a platform and a MATLAB/Simulink virtual model, the system can continuously represent the lack of simple, integrated, and cost-effective Digital Twin architectures specifically designed for aircraft-wing structural health monitoring with machine learning.

The proposed research attempts to address this gap by integrating relatively low-cost sensors, cloud communication, simulation, and machine learning into one monitoring system.

Main objectives

The study aims to:

1. Develop an IoT-based aircraft-wing prototype capable of detecting bending and vibration in real time.

2. Create a Digital Twin using MATLAB/Simulink.

3. Synchronize the physical wing and digital model through ThingSpeak.

4. Apply a Decision Tree classifier to identify structural deviations.

5. Support intelligent condition-based maintenance to improve aircraft safety and maintenance efficiency.

Literature review findings

The literature shows that Digital Twin technology has developed from conventional simulation into a broader cyber-physical technology involving physical systems, virtual models, data, connectivity, and intelligent services.

Previous studies have demonstrated applications in:

• Aircraft structural health monitoring

• Predictive and condition-based maintenance

• Aircraft lifecycle management

• Fatigue and crack-growth prediction

• Real-time flight monitoring

• Damage identification

• Aircraft assembly and quality control

• Aerodynamic and structural interaction

• Remaining useful life prediction

• Aircraft performance monitoring

• Anomaly detection

• Assembly visualization

Several approaches combine physics-based models with machine learning, while others use Bayesian methods, Kalman filtering, finite-element models, sensor fusion, and probabilistic modelling to deal with uncertainty.

Important insights from previous research

The literature particularly emphasizes that Digital Twins can:

• Continuously monitor aircraft structures instead of relying only on periodic inspections.

• Combine sensor measurements with physics-based models.

• Detect abnormalities and structural damage at an earlier stage.

• Estimate remaining useful life.

• Support aircraft-specific rather than fleet-average maintenance decisions.

• Reduce unnecessary maintenance.

• Improve traceability across the aircraft lifecycle.

• Assist technicians through real-time visualization and decision support.

However, important challenges remain, including sensor reliability, data integration, cybersecurity, interoperability, computational requirements, communication latency, uncertainty, model validation, scalability, and regulatory certification.

Overall conclusion

The text proposes a low-cost, real-time Digital Twin framework for aircraft-wing structural health monitoring. By connecting physical sensors to a cloud platform and a MATLAB/Simulink virtual model, the system can continuously represent the wing's condition. A Decision Tree classifier is then used to identify deviations between normal and abnormal structural behaviour.

Conclusion

In this paper, we present a Digital Twin framework for aircraft wing maintenance with machine learning-based deviation detection. The framework combines IoT sensing, cloud communication, virtual simulation, and intelligent structural health monitoring into a single cyber-physical framework. We propose a system for real-time visualisation of structural behaviour and predictive analysis through continuous synchronisation of the physical aircraft wing prototype with the virtual counterpart using ESP32, Flex Sensor, MPU6050, ThingSpeak, and MATLAB/Simulink. The developed Digital Twin, in contrast to traditional maintenance based on scheduling, continuously monitors the condition of the aircraft by capturing deformation and vibration data, updating the virtual aircraft wing, and classifying the condition of the structure through the Decision Tree classifier. Machine Learning adds value to maintenance by helping detect any deviations in the structure before catastrophic failure. The practical realisation confirms the low cost, scalability, and intelligence of the proposed approach as a solution for aircraft structural health monitoring. Indeed, the integration of physical sensing, cloud synchronisation, virtual modelling, and automatic anomaly detection can be considered effective for future applications in aerospace maintenance and digital twin-based smart aviation systems.

References

[1] B. Stanton, Digital Twin Technology. Atkins Global, Jan. 2017. [2] S. Haag and R. Anderl, “Digital Twin–Proof of Concept,” Manufacturing Letters, vol. 15, pp. 64–66, 2018. [3] Z. Bing et al., “Digital Twin: Concepts, Enabling Technologies, and Applications,” Journal of the Brazilian Society of Mechanical Sciences and Engineering, 2024. [4] C. Kan and C. J. Anumba, “Digital Twins as the Next Phase of Cyber-Physical Systems in Construction,” in ASCE Conference, 2019. [5] M. Mugundan, S. S. Sabithulla, S. Vignesh, A. Vijayalakshmi, and K. Virumandi, “Digital Twin Technology for Aircraft Wing Maintenance,” 2025. [6] C. M. Ezhilarasu, Z. Skaf, and I. K. Jennions, “A Generalised Methodology for the Diagnosis of Aircraft Systems,” IEEE Access, vol. 9, 2021. [7] X. Lai, W. Sun, L. Yang, X. He, Y. Pang, and X. Song, “Digital Twin-Based Structural Health Monitoring by Combining Measurement and Computational Data: An Aircraft Wing Example,” Journal of Manufacturing Systems, 2023. [8] W. Zhang et al., “FlightTwin: A Generalized Digital Twin Framework,” IEEE Access, 2024. [9] C. Li et al., “Dynamic Bayesian Network for Aircraft Wing Health Monitoring,” AIAA Journal, 2017. [10] L. Li et al., “Digital Twin in Aerospace Industry,” IEEE Access, 2022. [11] E. J. Tuegel, “Airframe Digital Twin Challenges,” in AIAA Conference, 2012. [12] Y. Liangliang et al., “Real-Time Acceleration Integration for Aircraft Wing Digital Twin,” 2024. [13] H. Cai et al., “Quality Deviation Control for Aircraft Using Digital Twin,” Journal of Computing and Information Science in Engineering, 2021. [14] T. Wang and Z. Liu, Digital Twin for Aircraft Maintenance. Springer, 2022. [15] M. Uzun et al., “Hybrid Digital Twin Flight Performance Model,” in IEEE Aerospace Conference, 2019. [16] C. C. Peng and Y. H. Chen, “UAV Rotary Engine Anomaly Detection via Digital Twin,” IEEE Transactions on Aerospace and Electronic Systems, 2024. [17] E. J. Tuegel et al., “Reengineering Aircraft Structural Life Prediction Using a Digital Twin Paradigm,” 2011. [18] C. Dong et al., “6-DOF Digital Twin for Fixed-Wing Aircraft,” Springer, 2025. [19] L. Chen et al., “Damage Identification Using Digital Twin,” Engineering Proceedings, 2025. [20] B. Liang et al., “Displacement Field Perception for Digital Twin,” Sensors, vol. 20, no. 18, 2020. [21] Q. Zhang et al., “Digital Thread-Based Digital Twin Framework,” Journal of Manufacturing Systems, 2022. [22] M. Liao et al., “Digital Twin for Aircraft Structural Lifecycle,” in SPIE Proceedings, 2023. [23] M. Li et al., “Assembly Visualization Using Digital Twin,” IEEE, 2024. [24] T. E. Makuni et al., “Digital Twin for Aero-Structural Interaction,” 2026. [25] G. M. Bisanti et al., “Digital Twin for Aircraft Maintenance: A Systematic Review,” Internet of Things, 2023.

Copyright

Copyright © 2026 R Kshamaa Rajoop, Ishanvi M H, Jesy Jeff Laura E, Harshada G, Mohan Babu G. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.