Book a demo

Cut patent&paper research from weeks to hours with PatSnap Eureka AI!

Try now

Acoustic Emission Delamination Detection — PatSnap Eureka

Acoustic Emission Delamination Detection — PatSnap Eureka
Structural Health Monitoring · Wind Energy

Acoustic Emission Monitoring for Early-Stage Delamination in Wind Turbine Blade Composites

Drawing on more than 50 patent and literature sources, this page explains how acoustic emission (AE) monitoring detects subsurface delamination in wind turbine blade composites before it becomes macroscopically visible — covering AE physics, signal processing, machine learning classification, and full-scale SHM deployments.

AE Delamination Detection Workflow: 5-Stage Process from Stress Wave Generation to SHM Alert Five-stage acoustic emission monitoring workflow for wind turbine blade delamination detection: (1) delamination releases stress waves, (2) piezoelectric sensors capture signals, (3) time-frequency processing, (4) ML damage classification, (5) SHM alert issued. Based on PatSnap Eureka analysis of 50+ sources. CRACK ONSET SENSOR ARRAY SIGNAL PROCESS ML CLASSIFY SHM ALERT Delamination stress wave PVDF / PZT piezo sensors Wavelet / time-freq k-NN / PCA 88% accuracy Pre-critical detection AE Delamination Detection Pipeline From subsurface crack onset to structural health alert Elastic stress wave propagation in composite laminate
50+
Patent & literature sources analysed (2009–2025)
88%
Damage classification accuracy via supervised k-NN model (Federal Univ. Santa Catarina, 2023)
52 m
Commercial blade length validated in Siemens Gamesa full-scale fatigue test (2022)
Polynomial degree correlating cumulative AE energy to delamination crack length (Amirkabir, 2017)
Physical mechanisms

How Delamination Generates Acoustic Emission Signals

Acoustic emission monitoring functions by detecting transient elastic stress waves released when damage nucleates or propagates within a material. In composite laminates such as those used in wind turbine blades, multiple competing damage mechanisms — matrix cracking, fiber/matrix debonding, fiber breakage, and delamination — each generate distinctive stress wave signatures. As demonstrated by INSA de Lyon/MATEIS (2021), finite element modelling (FEM) of AE waveforms resulting from fiber break and fiber/matrix debonding enables researchers to distinguish the physical origins of signals and better interpret real sensor measurements.

Delamination is particularly insidious because it develops subsurface and is invisible to visual inspection until catastrophic propagation occurs. The delamination crack growth process releases strain energy as it advances, and this energy is directly measurable via cumulative AE activity. Amirkabir University of Technology (2017) established a third-degree polynomial correlation between cumulative AE energy and crack growth length for glass/epoxy double cantilever beam (DCB) specimens under both quasi-static and fatigue loading, enabling predictive delamination prognosis well before structural failure.

Mode discrimination between in-plane (symmetric, S0) and out-of-plane (asymmetric, A0) Lamb wave modes provides an additional layer of delamination identification. The Measuring the Amplitude Ratio (MAR) approach from Fraunhofer Ernst-Mach-Institut (2011) demonstrated that the ratio between S0 and A0 Lamb wave modes changes systematically when delamination, rather than matrix cracking, is the dominant damage event. Research from Northeast Petroleum University (2022) further confirmed a one-to-one correspondence between AE parameter signatures and specific damage mechanisms at multidirectional ply interfaces (0°, 30°, 45°, and 60°) before macroscopic crack opening is observable. Standards bodies such as ASTM International and ISO provide the underpinning test frameworks referenced by these studies.

S0 / A0
Lamb wave mode ratio — changes systematically at delamination onset (Fraunhofer EMI, 2011)
4 plies
Ply orientations (0°, 30°, 45°, 60°) at which AE resolves delamination onset individually (NE Petroleum Univ., 2022)
DCB
Double cantilever beam test geometry used to quantify AE energy vs. crack length correlation
FEM
Finite element modelling confirms characteristic frequency and amplitude content per damage mode
  • Matrix cracking: high-frequency, low-amplitude AE bursts
  • Fiber/matrix debonding: intermediate frequency, moderate amplitude
  • Delamination: lower-frequency content, higher energy release
  • Fiber breakage: highest amplitude, broadband frequency
  • Friction on crack faces: indirect indicator of pre-existing delamination
Data visualisation

AE Classification Accuracy and Technology Readiness

Key quantitative benchmarks from the 50+ source dataset, spanning signal processing method accuracy and deployment readiness levels across application contexts.

AE Damage Classification Accuracy by Signal Processing Method

Supervised k-NN achieves the highest reported accuracy (88%) for delamination identification in commercial composites; unsupervised and wavelet methods provide qualitative discrimination.

AE Damage Classification Accuracy by Method: k-NN Supervised ML 88%, Wavelet Clustering pre-macroscopic detection, k-Means PCA unsupervised delamination isolation, MAR Lamb Wave delamination vs matrix cracking discrimination Bar chart comparing acoustic emission signal processing methods for composite delamination classification. The supervised k-nearest neighbours model from Federal University of Santa Catarina (2023) achieved 88% classification accuracy — the highest quantified value in the reviewed dataset. Source: PatSnap Eureka analysis of 50+ patent and literature sources (2009–2025). 100% 75% 50% 25% 0% 88% k-NN Supervised pre-macro Wavelet Clustering isolated k-Means + PCA discriminated MAR Lamb Wave

AE Monitoring Technology Readiness by Deployment Context

From laboratory DCB specimens (highest readiness) to in-service offshore SHM (active research frontier), showing the progression of AE validation across contexts.

AE Technology Readiness by Context: Lab DCB/coupon TRL 9, Component fatigue testing TRL 7, Full-scale 52m blade fatigue TRL 6, In-service onshore SHM TRL 5, In-service offshore SHM TRL 4 Horizontal bar chart illustrating technology readiness levels (TRL, 1–9 scale) for acoustic emission delamination monitoring across five deployment contexts, from controlled laboratory specimens to operational offshore wind turbines. Full-scale blade validation by Siemens Gamesa (2022) anchors TRL 6. Source: PatSnap Eureka analysis. Lab DCB / coupon Component fatigue Full-scale 52m blade In-service onshore In-service offshore TRL 9 TRL 7 TRL 6 TRL 5 TRL 4 TRL 1 TRL 9 ← Siemens Gamesa 2022

Search 50+ AE delamination patents and studies — filtered by assignee, date, or technique.

Search AE Patents in Eureka →
Signal processing & classification

Translating Raw AE Waveforms into Delamination Diagnostics

Time-frequency analysis, wavelet transforms, and machine learning clustering are the primary strategies for converting AE sensor data into actionable delamination identification, as documented across the reviewed dataset.

Unsupervised ML

k-Means Clustering and PCA for Automated Damage Separation

Unsupervised pattern recognition methods including k-means clustering and Principal Component Analysis (PCA) enable automated separation of AE events by damage type without requiring prior labeled datasets. VŠB—Technical University of Ostrava (2021) demonstrated a two-step technique combining short-time frequency spectra with boundary curve filtering that successfully isolated delamination-related events from fiber fracture signals during three-point bending tests. The boundary curve was calibrated against tensile tests on bare fiber bundles.

Transferable to blade composite architectures
Supervised ML

k-NN Model Achieves 88% Classification Accuracy

Federal University of Santa Catarina (2023) trained a k-nearest neighbors (k-NN) model using fingerprint AE data collected from composite constituents (fiber, matrix, and interface) tested individually, achieving 88% classification accuracy when applied to tensile tests of commercial composites. The model provided localization, timing, frequency, and intensity information for each damage mode. UC Santa Barbara (2021) established foundational requirements for physics-informed waveform feature extraction combined with neural network clustering strategies.

88% accuracy on commercial composites
Wavelet Analysis

Temporal Resolution for Sequential Delamination Emergence

Kielce University of Technology (2017) demonstrated that wavelet decomposition of AE signals during tensile loading of CFRP specimens enabled detection and classification of debonding, matrix cracking, and delamination as distinct temporal events. Delamination signals, characterised by lower-frequency content and higher energy release compared to matrix cracking, were identifiable prior to any macroscopic specimen degradation. Semnan University (2021) extended these findings to cyclic fatigue regimes relevant to operational wind turbine loading using fuzzy and wavelet clustering methods.

Pre-macroscopic delamination identification
Amplitude & Energy Thresholds

Hybrid Composite Delamination Separation via AE Amplitude

University of Bristol (2018) found that in SGlass/TR30-Carbon hybrid laminates, low-amplitude AE signals corresponded to carbon/glass interface delamination while high-amplitude signals were linked to carbon layer fragmentation. The study further noted that friction-related AE signals during unloading cycles provide indirect evidence of pre-existing delamination crack faces in contact — an important early indicator for blade fatigue monitoring. Delft University of Technology's 2020 review confirms AE is valued for its high sensitivity and capacity for damage initiation detection across large monitored areas.

Friction signals = indirect delamination evidence
PatSnap Eureka

Map the full AE signal processing patent landscape

Identify which institutions hold IP in wavelet, ML, and MAR-based AE classification for composite SHM.

Analyse AE Classification Patents →
SHM implementations & patents

Key Organisations, Patents, and Validated Deployments

The dataset reveals a focused set of organisations each contributing distinct technical advances, from full-scale blade fatigue testing to active patent families and offshore diagnostic frameworks.

Organisation Key Contribution Application Context IP / Publication Status
Siemens Gamesa Renewable Energy Full-scale fatigue testing of 52-meter commercial blade; AE detected delamination propagation before strain gauges or accelerometers registered changes. Active EP patent on electrically conductive fastener-based delamination surveillance (2021). Full-scale blade, operational Active Patent
University of South Carolina Active US patent family protecting AE-based damage ascertainment methods for impacted composites. Separates high-frequency damage initiation AE from low-frequency flexural wave components. Applications filed 2021, 2024, and 2025. Impacted composites, in-situ blade monitoring 3 Active/Pending US Patents
Technical University of Denmark (Risø) Mechanism-based SHM framework mapping surface erosion, adhesive fatigue, laminate cracking, plydrop delamination, and compressive kinking to tailored AE monitoring strategies. Plydrop delamination detectable under static and dynamic loading. Operational wind turbine blades Peer-Reviewed (2020)
Windhunter Group (Poland) Coherence function analysis of AE signals for offshore blade diagnostics. Non-linear wave phenomena associated with delamination distinguishable from undamaged blade responses using inverse transformation spectral analysis under operational conditions. Offshore wind turbines, in-service Peer-Reviewed (2023)
🔒
Unlock the full organisation & patent comparison
See Cardiff University, Amirkabir, Fraunhofer, Cranfield, and Delft entries — including Bayesian localization methods, embedded sensor reviews, and predictive maintenance correlations.
Cardiff Bayesian SHM Amirkabir AE-crack correlation Fraunhofer MAR method + more
View Full Dataset in Eureka →

Track active AE patents across all assignees in real time

PatSnap Eureka monitors new filings, continuations, and status changes across the full AE composite monitoring IP landscape.

Monitor AE Patent Activity →
Key takeaways

Seven Validated Findings from the AE Delamination Research Dataset

Each finding below is directly traceable to a named source within the 50+ patent and literature dataset (2009–2025).

📡

AE Detects Delamination Before Macroscopic Visibility

Full-scale fatigue testing of a 52-meter commercial blade (Siemens Gamesa, 2022) demonstrated that AE sensors registered damage propagation from laminate cracks into delaminations before strain gauges or accelerometers registered equivalent changes, confirming AE's superior sensitivity in the pre-critical damage phase.

📈

Cumulative AE Energy Correlates with Crack Length via 3rd-Degree Polynomial

A quantitative third-degree polynomial relationship between cumulative AE energy and crack growth was established for glass/epoxy DCB specimens under both static and fatigue loading (Amirkabir University of Technology, 2017), enabling predictive delamination prognosis well before structural failure.

〰️

Lamb Wave Mode Ratios Enable Delamination-Type Discrimination

The Measuring the Amplitude Ratio (MAR) approach comparing symmetric (S0) and asymmetric (A0) Lamb wave modes allows differentiation of delamination from matrix cracking events in plate-like composite structures (Fraunhofer Ernst-Mach-Institut, 2011).

🤖

Supervised k-NN Achieves 88% Damage Classification Accuracy

A k-NN model trained on constituent-level AE fingerprints achieved 88% damage classification accuracy in commercial composites (Federal University of Santa Catarina, 2023), demonstrating practical feasibility for automated delamination detection in wind turbine blade inspection.

🔒
Unlock 3 additional validated findings
Including plydrop delamination site targeting, offshore coherence function diagnostics, and embedded PVDF/PZT smart blade architectures.
Plydrop SHM targeting Offshore coherence analysis Smart embedded sensors
Read All Findings in Eureka →
Innovation trends

AE Patent Activity and Publication Volume (2009–2025)

The dataset spans over 50 sources, with activity concentrated among six dominant assignees and a clear evolution from parameter-based to machine learning-integrated AE monitoring.

AE Patent & Publication Activity by Organisation Type (2009–2025)

Industrial assignees (Siemens Gamesa) anchor full-scale validation; academic institutions drive signal processing and ML innovation; emerging groups target offshore and embedded sensor architectures.

AE Patent and Publication Activity by Organisation: Siemens Gamesa (industrial, active EP patent 2021 + 52m blade study 2022), University of South Carolina (3 US patent families: 2021, 2024, 2025), DTU Risø (mechanism SHM framework 2020), Delft University (comprehensive review 2020), Windhunter Group (offshore coherence analysis 2023), Cardiff University (Bayesian localization 2017) Timeline and activity map showing acoustic emission patent filings and key publications from dominant assignees spanning 2009 to 2025. University of South Carolina shows the most active IP expansion with three patent family filings across 2021–2025. Source: PatSnap Eureka dataset of 50+ sources. 2009 2011 2015 2017 2020 2022 2025 Siemens Gamesa Univ. S. Carolina DTU Risø Delft / Fraunhofer Windhunter / Cardiff 22 EP 21 24 25 2020 11 20 17 2023 Parameter-based → ML-integrated AE monitoring trend →

Identify white spaces and emerging players in AE composite monitoring IP.

Explore the Full Patent Landscape →
Embedded sensors & offshore deployment

Smart Blade Architectures and Offshore AE Diagnostics

Embedded piezoelectric sensors — including thin PVDF films and PZT patches — are increasingly being integrated directly into composite layups, as reviewed by Cranfield University (2023) in their comprehensive review of AE testing for polymer-based composite SHM. This approach enables smart blade structures capable of autonomous, continuous delamination surveillance without external sensor attachment, aligning with the broader direction in composite structural health monitoring.

For offshore wind turbines specifically, Windhunter Group (2023) proposed a coherence function analysis approach for blade condition monitoring. The authors found that blade damage involving delamination manifests as non-linear wave phenomena in AE signals that are absent in the harmonic components of undamaged blade responses. Using inverse transformation in the signal analysis process, they demonstrated the ability to distinguish damaged from undamaged blade spectral signatures under operational conditions — a critical requirement for in-service offshore deployment where manual inspection is costly and hazardous.

Sensor placement optimisation remains an open challenge identified by Delft University of Technology (2020), alongside signal attenuation in thick laminates. The National Renewable Energy Laboratory (NREL) and IRENA have both highlighted blade structural integrity monitoring as a priority for reducing offshore wind operations and maintenance costs. University of Castilla-La Mancha (2016) established spatial resolution capabilities using macro-fiber composite (MFC) sensors in triangular arrays with time-difference-of-arrival (TDOA) source location algorithms for crack position identification on blade cross-sections. PatSnap's advanced materials intelligence platform enables tracking of embedded sensor material innovations across the composite SHM IP landscape.

PVDF
Thin film piezoelectric sensor type integrated directly into composite layups for autonomous AE monitoring
PZT
Piezoelectric patch sensor type embedded in blade composites without significantly compromising structural properties
MFC
Macro-fiber composite sensors arranged in triangular arrays with TDOA algorithms for crack location (Castilla-La Mancha, 2016)
3 sensors
Minimum MFC sensor count positioned on blade cross-section for TDOA-based crack location
Open Challenges (Delft, 2020)
  • Signal attenuation in thick laminates
  • Sensor placement optimisation
  • Separating operational vibration from damage AE
  • Uncertainty quantification for industrial deployment
Frequently asked questions

Acoustic Emission Delamination Detection — Key Questions Answered

Still have questions about AE delamination monitoring? Let PatSnap Eureka search the patent and literature database for you.

Ask Eureka About AE Delamination →
PatSnap Eureka

Detect Delamination Earlier — Backed by the Full AE Patent & Literature Landscape

Join 18,000+ innovators already using PatSnap Eureka to accelerate their R&D. Search 50+ AE monitoring sources, track active patents from Siemens Gamesa and the University of South Carolina, and map emerging embedded sensor technologies — all in one AI-native platform.

References

  1. Acoustic Emission Signal Due to Fiber Break and Fiber Matrix Debonding in Model Composite: A Computational Study — INSA de Lyon / MATEIS UMR 5510, 2021
  2. The Use of Coherence Functions of Acoustic Emission Signals as a Method for Diagnosing Wind Turbine Blades — Windhunter Group, Poland, 2023
  3. Structural health monitoring of 52-meter wind turbine blade: Detection of damage propagation during fatigue testing — Siemens Gamesa Renewable Energy, 2022
  4. Surveillance of delamination propagation in a composite structure in a wind turbine component — Siemens Gamesa Renewable Energy A/S, EP patent, active, 2021
  5. Damage Mechanism Based Approach to the Structural Health Monitoring of Wind Turbine Blades — Technical University of Denmark, Risø Campus, 2020
  6. Acoustic Emission-Based Methodology to Evaluate Delamination Crack Growth Under Quasi-static and Fatigue Loading Conditions — Amirkabir University of Technology, 2017
  7. Characterisation of Damage in Composite Structures using Acoustic Emission — Fraunhofer Institut für Kurzzeitdynamik, Ernst-Mach-Institut, 2011
  8. Damage characterization of laminated composites using acoustic emission: A review — Delft University of Technology, 2020
  9. Acoustic emission method to ascertain damage occurrence in impacted composites — University of South Carolina, US patent, active, 2021
  10. Acoustic emission method to ascertain damage occurrence in impacted composites — University of South Carolina, US patent, active, 2024
  11. Acoustic emission method to ascertain damage occurrence in impacted composites — University of South Carolina, US patent, pending, 2025
  12. A Review on Acoustic Emission Testing for Structural Health Monitoring of Polymer-Based Composites — Cranfield University, 2023
  13. Identifying damage mechanisms of composites by acoustic emission and supervised machine learning — Federal University of Santa Catarina, Brazil, 2023
  14. Damage mechanism identification in composites via machine learning and acoustic emission — University of California Santa Barbara, 2021
  15. Damage Analysis of Composite CFRP Tubes Using Acoustic Emission Monitoring and Pattern Recognition Approach — VŠB—Technical University of Ostrava, 2021
  16. Wavelet Analysis of Acoustic Emissions during Tensile Test of Carbon Fibre Reinforced Polymer Composites — Kielce University of Technology, 2017
  17. Prediction of fiber breakage and matrix cracking in polymeric composites under low-cycle fatigue regimes by fuzzy and wavelet clustering of acoustic emission signals — Semnan University, 2021
  18. Acoustic emission monitoring of thin ply hybrid composites under repeated quasi-static tensile loading — University of Bristol, 2018
  19. Study on Delamination Damage of CFRP Laminates Based on Acoustic Emission and Micro Visualization — Northeast Petroleum University, 2022
  20. A New Fault Location Approach for Acoustic Emission Techniques in Wind Turbines — University of Castilla-La Mancha, 2016
  21. Acoustic emission based damage localization in composites structures using Bayesian identification — Cardiff University, 2017
  22. National Renewable Energy Laboratory (NREL) — Wind turbine structural integrity and blade monitoring research
  23. International Renewable Energy Agency (IRENA) — Offshore wind operations and maintenance cost reduction priorities
  24. ASTM International — Standard test methods for acoustic emission monitoring of composites
  25. ISO — International standards for non-destructive evaluation and structural health monitoring

All data and statistics on this page are sourced from the references above and from PatSnap's proprietary innovation intelligence platform. Patent and literature dataset spans 2009–2025, comprising 50+ sources analysed via PatSnap Eureka.

Ask PatSnap Eureka
Ask PatSnap Eureka
AI innovation intelligence · always on
Ask anything about acoustic emission delamination monitoring.
PatSnap Eureka searches patents and research to answer instantly.
Try asking
Powered by PatSnap Eureka