Structural health monitoring (SHM) in fiber-reinforced polymer (FRP) composites is essential to ensure safety and reliability during service, particularly in critical industries such as aerospace and wind energy. Traditional methods of analyzing Acoustic Emission (AE) signals in the time domain often fail to accurately detect subtle or early-stage damage, limiting their effectiveness. The present study introduces a novel approach that integrates frequency-domain analysis using the fast Fourier transform (FFT) with deep learning techniques for more accurate and proactive damage detection. AE signals are first transformed into the frequency domain, where significant frequency components are extracted and used as inputs to an autoencoder network. The autoencoder model reduces the dimensionality of the data while preserving essential features, enabling unsupervised clustering to identify distinct damage states. Temporal damage evolution is modeled using Markov chain analysis to provide insights into how damage progresses over time. The proposed method achieves a reconstruction error of 0.0017 and a high R-squared value of 0.95, indicating the autoencoder’s effectiveness in learning compact representations while minimizing information loss. Clustering results, with a silhouette score of 0.37, demonstrate well-separated clusters that correspond to different damage stages. Markov chain analysis captures the transitions between damage states, providing a predictive framework for assessing damage progression. These findings highlight the potential of the proposed approach for early damage detection and predictive maintenance, which significantly improves the effectiveness of AE-based SHM systems in reducing downtime and extending component lifespan.

This study presents a novel methodology for the real-time characterisation and quantitative assessment of damage in fibre-reinforced polymers (FRPs) using acoustic emission (AE) techniques. While FRPs offer superior mechanical properties for structural applications, their anisotropic nature introduces complex damage mechanisms that are challenging to detect with conventional inspection methods. Our approach advances beyond traditional peak frequency analysis by implementing a multi-variant frequency assessment that can detect and evaluate simultaneously occurring damage modes. By applying the fast Fourier transform and examining multiple frequency peaks within AE signals, we successfully identified five distinct damage mechanisms in carbon fibre composites: matrix cracking (100–200 kHz), delamination (205–265 kHz), debonding (270–320 kHz), fibre fracture (330–385 kHz), and fibre pullout (395–490 kHz). A comparative analysis with wavelet transform methods demonstrated that our approach provides earlier detection of critical damage events, with delamination identified approximately 28 s sooner than with conventional techniques. The proposed methodology enables a more accurate quantitative assessment of structural health, facilitating timely maintenance interventions for large-scale FRP structures, such as wind turbine blades, thereby enhancing reliability while reducing operational downtime and maintenance costs.

The accurate evaluation of the structural health of in-service fibre-reinforced polymer components is particularly interesting. Acoustic emission has been widely studied for monitoring damage initiation and propagation in such materials. However, the accurate identification of exact damage modes together with the accurate quantification of damage severity through analysis of acoustic emission signals remain challenging. The present study explores the applicability of frequency-based multi-variant analysis for identifying damage modes and quantifying the level of sustained damage in fibre-reinforced composites, highlighting the limitations of existing methods.

Structural health monitoring (SHM) in Composite Fiber Reinforced Polymer (CFRP) structures is essential to ensure safety and reliability during service, particularly in critical industries such as aerospace and wind energy. Traditional methods of analysing Acoustic Emission (AE) signals in the time domain often fail to accurately detect subtle or early-stage damage, limiting their effectiveness. The present study introduces a novel approach that integrates frequency-domain analysis using Fast Fourier Transform (FFT) with deep learning techniques for more accurate and proactive damage detection. AE signals are first transformed into the frequency domain, where significant frequency components are extracted and used as input for an autoencoder network. The autoencoder reduces the dimensionality of the data while preserving essential features, enabling unsupervised clustering to identify distinct damage states. Temporal damage evolution is modeled using Markov chain analysis to provide insights into how damage progresses over time. The proposed method achieves a reconstruction error of 0.0017 and a high R-squared value of 0.95, indicating the autoencoder's effectiveness in learning compact representations while minimising information loss. Clustering results, with a silhouette score of 0.37, demonstrate well-separated clusters that correspond to different damage stages. Markov chain analysis captures the transitions between damage states, providing a predictive framework for assessing damage progression. These findings highlight the potential of the proposed approach for early damage detection and predictive maintenance, significantly improving the effectiveness of AE-based SHM systems in reducing downtime and extending component lifespan.

Renewable energy sources are expanding rapidly, with wind energy representing one of the fastest-growing sectors. In parallel, digital twins are transforming structural monitoring and inspection by enabling near real-time structural assessment. However, the high computational cost associated with high-fidelity numerical models requires reduced-order or surrogate modelling strategies. Within this context, composite laminate thickness plays a critical role in wind turbine blade design and assessment. In this study, a computational fluid dynamics (CFD)-based numerical model developed in ANSYS was employed to predict von Mises stresses in a three-dimensional wind turbine blade geometry derived from the Carbo4Power project. The numerical results were validated against published literature, enabling identification of a thickness–stress relationship. This foundational analysis supports the development of reduced-order models suitable for digital twin applications in wind turbine blade structural assessment.