Wind turbine blade inspection methods
Wind Turbine Drone Inspection vs Blade Robot
A neutral planning guide for deciding what exterior aerial imagery can answer, what requires access inside the blade, and how to specify a traceable evidence package without turning one method into a universal replacement for another.
Decision first
Start with the evidence question.
Choose the method after defining the surface and decision. “Wind turbine blade inspection” can refer to an exterior visual survey, an internal cavity survey, a close manual examination or a specialised NDT procedure. Those scopes are not interchangeable simply because each can produce images.
A current peer-reviewed review of drone-based blade inspection describes aerial platforms as part of a wider chain that includes image acquisition, defect detection and analysis. That chain is useful for accessible exterior surfaces, but image evidence alone does not establish what is happening on an unseen internal surface or within a laminate.
Write one sentence before selecting equipment.
Example: “Locate and retain reviewable images of visible internal surface features along the accessible pressure-side cavity, referenced by blade, cavity and distance from root.” This is more actionable than “inspect the blade with AI.”
Separate four questions in the inspection brief
- Surface: exterior shell, accessible internal cavity, bond line, root area or another defined zone.
- Phenomenon: visible surface feature, dimensional change, thermal anomaly, crack-like indication or subsurface discontinuity.
- Evidence: raw image, video, location reference, annotation, measurement, NDT data or signed report.
- Decision: screen, trend, investigate, repair-plan or assess against an approved acceptance criterion.
Neutral comparison
Compare by access boundary, not novelty.
Each method begins with a different physical access boundary. The table is a planning aid, not a performance ranking. Actual capability depends on the chosen platform, sensor, procedure, operator, asset geometry and operating conditions.
| Method | Primary evidence | Access boundary | Useful when | Plan explicitly |
|---|---|---|---|---|
| Exterior drone | Exterior stills or video, sometimes with thermal or other supported sensors | Line of sight to the outer blade surface | A repeatable exterior visual record is the required first screen | Weather, stand-off distance, image scale, angle, lighting, regulation and missed-surface rules |
| Internal crawler | Close internal imagery tied to an accessible crawler path | Entry opening, cavity geometry, adhesion surface and traversable path | The decision requires visible evidence from inside an accessible cavity | Insertion, recovery, obstacles, tether, location reference, lighting and coverage statement |
| Internal UAV prototype | Internal imagery from a flying platform | Opening size, confined flight volume and recoverable navigation envelope | A validated platform can fly safely in the defined cavity | GPS-denied navigation, airflow, collision clearance, lighting, battery and retrieval |
| Rope-access visual | Close exterior human observation and selected images | Reachable external areas under an approved access and rescue plan | Hands-on confirmation or another close task is required | Safety controls, weather window, positioning, repeatability and record format |
Internal UAV work is not hypothetical, but it should not be conflated with routine exterior drone inspection. A 2021 experimental study of UAV inspection inside a blade describes the distinct navigation and image-capture problem in a confined, dark environment.
Crawler configuration
Plan what the internal record must retain.
An internal crawler is valuable only when its captures remain interpretable after the field visit. At minimum, each observation needs the source image, blade and cavity identity, a location reference, capture conditions, review state and the rule used for any classification.
The Fraunhofer RIWEA project illustrates the broader principle: robotic positioning, sensors and a coordinate-based condition log form one inspection system. RIWEA is an external robotic research system—not evidence for HZBR-60Pro—but its documentation reinforces why capture and position should stay connected.
- 01Confirm physical access
Record opening dimensions, cavity, webs, known obstructions, surface condition and the planned recovery method.
- 02Define the path
State the start reference, intended travel direction, accessible zones and how skipped or blocked areas will be reported.
- 03Specify capture quality
Agree required views, lighting, focus, image resolution, repeat-capture triggers and retention format before mobilisation.
- 04Keep human review visible
Separate raw capture, AI-assisted annotation, engineering classification and the asset owner's final action.
HZBR-60Pro product-document boundary
The published 70 m / ≤60 minute scenario applies to one cavity over a 70 m internal area in a 95 m blade, with the documented setup and an unobstructed robot path. The separate “up to 70%” coverage value remains configuration-dependent. Review the HZBR-60Pro specifications and conditions.
Evidence architecture
Use one location model across methods.
A combined inspection becomes useful when exterior and internal findings can be correlated. Use a shared asset identifier, blade designation, span or distance reference, surface/cavity label, timestamp and image identifier. Without that structure, two large image sets can still leave an engineer unable to compare observations.
A practical combined sequence
- Screen the exterior. Capture the approved outer-surface zones and log every unobserved or low-confidence area.
- Target the internal scope. Use asset history, exterior observations and known blade geometry to define the cavity path—without assuming a visible exterior feature has an internal counterpart.
- Escalate the method. Where the question concerns a bond, laminate or hidden discontinuity, select an approved NDT procedure rather than inferring it from visual images.
- Retain the decision trail. Keep original captures, annotations, reviewer identity, classification basis and next action as separate fields.
The sample internal blade inspection report shows one way to separate source imagery, observation, classification and asset-owner decision. It is an illustrative information structure, not a customer case or diagnostic template.
Questions to put in an RFQ
- Which exact surfaces and zones are included, conditionally included or excluded?
- How are inaccessible, obstructed or poor-quality areas represented in the report?
- What is the smallest reviewable image feature at the planned distance and lighting?
- How is location established, and what is its stated uncertainty?
- Are AI outputs suggestions, classifications or final decisions—and who verifies them?
- Which raw files and metadata are delivered for later engineering review?
- What clock boundaries and asset conditions apply to any duration claim?
Evidence register
Sources and limitations
This guide combines method context from research and official project pages with clearly separated WindInspectTech product-document values.
- Electronics (2025): Review of drone-based wind-turbine blade inspection — peer-reviewed method overview for aerial acquisition and analysis.
- Energies (2021): Internal blade inspection using UAVs — experimental research on the distinct confined-space problem.
- Fraunhofer IFF: RIWEA automated rotor-blade inspection — official external robotic research-project description.
- BAM: InInspekt internal blade inspection project — official research-project context for autonomous internal inspection.
- Supplied product manual: 叶片内部巡检机器人 HZBR-60Pro 产品说明书, Version A, document
HZSN/BL-IR-001/2026— §§2.1–2.2, §3, §5, §6 and §§7.3.2–7.3.3. The manual pages identify 北京汇众数能科技有限公司, and §2.1 describes the crawler as a 汇众数能 product. See the translated HZBR-60Pro product specification and conditions.
Limitation: the external sources provide method context and do not independently validate HZBR-60Pro, its coverage, duration, image quality or suitability for a particular blade. A project-specific access and procedure review remains necessary.
See the source and editorial policy for how supplied product manuals, external method sources and claim limitations are kept separate.
Decision support
Frequently asked questions
Can a drone inspect inside a wind-turbine blade?
Research prototypes have demonstrated internal UAV inspection, but that is a different operating problem from a standard exterior drone survey. Internal flight must account for confined geometry, lighting, navigation without GPS, collision clearance and recoverability. Confirm whether a proposed system is designed and validated for the specific cavity.
Does an internal blade robot inspect every internal surface?
No universal coverage should be assumed. Openings, cavity geometry, webs, obstructions, surface condition, adhesion, tether management and the approved path all affect access. HZBR-60Pro documentation states up to 70% internal coverage, but that product-document value is not a guarantee for every blade.
Is drone inspection faster than an internal crawler?
There is no defensible universal answer. Time depends on turbine state, weather, access, mobilisation, blade geometry, capture scope, repeat images and reporting requirements. Compare a supplier's stated clock start and stop points, exclusions and rework policy before comparing duration figures.
Should an asset owner use both exterior and internal inspection?
Use both when the engineering decision needs evidence from both sides of the blade shell. Exterior imagery can document accessible outer surfaces; an internal crawler can document accessible cavity surfaces. The inspection plan should still identify any bond line, laminate or subsurface questions that require another NDT method.
Engineering review
Define the surfaces and evidence your decision needs.
Send the blade type, access opening, known constraints, required image references and reporting decision. We’ll identify which parts of the workflow need configuration review.