Airborne inspection and quality assurance

Intelligent airborne inspection

The standard inspection method for quality assurance of the exterior surfaces of wind turbines is to employ people trained as rope access or industrial climbers, but this is not the only way.

Drones are capable of undertaking visual inspection of turbines; supporting software and archiving can be used to analyse the data. While drones and airborne access will never be a complete substitute for manned inspection, and there will always be the need for rope access workers, there are many advantages to drone-based airborne inspection together with software analysis and archiving.

Aero Enterprise GmbH, a young company located in Linz, Austria, is active in the field of airborne inspection and quality assurance. The company was founded in 2013 and provides its service with a comprehensive system consisting of a helicopter-type flight-robot (drone) called AERO-SensorCopter, and a client-based AERO-Software package, consisting of a free licence (AERO-View) and a fee-based licence (AERO-Lyse). The analysis and reporting software is supported by a database in the background. Data mining and machine learning tools enable clients to use the data not only for statistics about the present state of the turbine but to predict upcoming demand for maintenance.

Customers are operators of wind farms, service companies, insurance companies and original equipment manufacturers, as well as technical experts in this field. With primary focus on the wind energy market, airborne inspection technology can also be applied to all other kinds of vertical objects; it also helps to reduce maintenance and long term after-sales costs.

Acquisition of data

With the proprietary AERO-SensorCopter, the rotor blades, nacelle and spinner of a turbine can be inspected visually. On request the drone can take a closer look at welding seams or at the transition area from concrete to steel on hybrid towers. During the company’s planned next step into the offshore market, inspection of the foundation will also be taken into consideration.

The electronically stabilised and gimballed sensor head is capable of sweeping up 70 and down 90 degrees each way and is equipped with two synchronised camera systems running in parallel, providing infrared and high-resolution images.

In the process of completing an airborne inspection a theoretical resolution of up to 12 pixels per square millimetre of the object can be achieved with the 42 megapixel camera and lens, but in practice the achievable resolution can vary depending on distance, illumination and vibration (e.g. caused by strong gusts of wind). The infrared camera provides real-time temperature data from the object, which can give extra information about the situation below the surface. This is very interesting when looking at the bonding areas of the upper and lower outer shell of the blade along the main beam.

The modular concept allows other payloads to be launched with the system, either standalone or partially integrated into the flight system (autopilot). Except for launching and landing, the whole flight process is fully automated and the drone generally flies close to the object at a distance between 5-10m.

Due to the helicopter design, the airborne inspection can be done in wind speed conditions up to 14m/s (50km/h or 27 knots). For best results the rotor system first has to be set so that the blades are roughly in a Mercedes-star position, with one blade looking straight up (no bolting is needed, only braking action). Then a complete flight along the blades from all four sides is performed.

Analysis of data

After some special post-processing works on the gathered data, the whole data package is uploaded to the database. Now the infrared and normal image data collected are displayed in Aero Enterprise’s self-developed, client based analysis software, AERO-Lyse. The customer, the dedicated expert or Aero Enterprise itself can now identify, classify and document the existing damaged areas visually with support of a 3D environment for better orientation. The damaged areas can easily be marked with the help of pop-up windows and menus, providing pre-selections and standardised sentences with respect to the selected damage. Information on dimensions and area sizes of anomalies are also available instantly. With the help of infrared pictures and passive thermography extra information from temperature differences on the surface is available and this can be used, for instance, to detect hot spots which may be an indication of a delamination of the outer shell against the main beam.

When the analysis reaches a point on the object where a damaged area was found from an earlier airborne inspection, a message comes up asking if the user would like to compare the past and present data. With this change in detection, which is possible because of the interaction of the front-end with the database in the background, consistent documentation is assured.

Finally, with the click of a button an automated report in pdf format can be generated. The whole process from data acquisition to a ready to use report takes only two to three working days.

On request, it is also possible to produce filtered data or reports (e.g. only the pictures with damage information needed for the customer’s enterprise resource planning (ERP) or interface could be provided).

Storage of data

The database stores airborne inspection data as a backup, but also, with the implementation of data mining and machine learning, can provide statistics and deep knowledge about future demand for predictive maintenance. The database is also the source for customisation of different classification standards or changes in language on the front-end software. Stored data is also the link which closes the quality cycle at the next inspection period, where old and new data can be compared. Most of the earlier human-based subjective interpretation of data can be ignored as the system provides a more objective statement. We call the first flight on the turbine the ‘digital birth certificate’.

Advantages of airborne inspection

Time and costs

One of the biggest advantages of this system lies in the saving of time for the inspection. The positioning of the ground station and the wind turbine rotor, calibration, airborne inspection and data transfer takes about two hours in total. The complete scan of the rotor system, nacelle and spinner takes only 25-40 minutes. This is significantly faster than a review by industrial climbers, especially for offshore turbines. As a result, the downtime for a wind turbine can be reduced on average by approximately 3 hours onshore or up to 14 hours offshore. Depending on distances, two to four wind turbines can be inspected per day.

Quality and traceability

With a repeated airborne inspection, a continuous documentation of the system status can be made, since the recorded images can be viewed and compared at any time. With the exact determination of the damage locations, the year-by-year changes in points affected are monitored. The extra digital information and data on earlier (or already known) problems on the blades, which can be related to factors such as production year, batch, site location or wind levels, increases the type and level of evaluation quality and improves traceability.

Accessibility and efficiency

The ability to inspect the rotor systems of turbines faster and the more convenient collection and evaluation of data makes the whole process more efficient and comparable. In addition, the capability to operate at wind speeds up to 14m/s and a minimum flight time of 30 minutes opens extra room for accessibility, especially in the time-consuming offshore business.

Technology

With the digitalised collection of data, it is easy to connect different work steps to other automated processes. Not only can the collected airborne inspection data interface with the customer’s ERP system but there is also the potential to connect the visual-based knowledge with vibration and metrological-based information to generate deep system knowledge and provide the next step into the future in the direction of deep learning and big data.

Health and safety and planning

The weather may limit when humans can undertake inspections. But there are also other limiting human factors like subjective decisions, work time frames, illness, simple mistakes or even casualties or fatalities which can cause a chain of inefficiency in planning.

Although no one likes to mention it, inspections by climbers may cause damage to the structure because of inattention or accidents. We have seen a lot of damage obviously caused by industrial climbers such as bent and chopped vortex generators damaged by someone stepping on them or slightly loose gurney flaps ripped off with the rope during the climb along the blades. The bigger and more efficient turbines become in the future, the more sophisticated their aerodynamic features and attachments. Think of the harm it could cause (to both the person and the structure) when a rope worker has to fight with the wind and hits against the dino-tails at the edge of a rotor blade.

Finally, there are costs in the form of cash, time or even reputation. An airborne inspection can help increase the yearly maintenance output and simultaneously helps to minimise the physically exhausting hands-on work on the blades in harsh weather conditions.

Drones and airborne access alone will never be a complete substitute for manned inspection and there will always be the need for rope workers. But airborne inspection can be a cost-reducing and supporting aftersales element to provide quality assurance over the lifetime of a plant.

However, to benefit from the advantages a change in paradigm may be needed, as the wind energy sector can sometimes be very conservative and not always open to innovation. Future orientated companies who are open for new developments are welcome to contact us. We are looking forward to testing partners and further potential cooperation partners in both business fields, onshore and offshore wind.

What is the partner programme?

The partner model is based on a division of labour between you as a co-operation partner and Aero Enterprise. The on-site inspection will not be conducted by Aero Enterprise, but by you with the help of your own conventional DJI consumer drone.

Aero Enterprise assists you in finding a recommended drone and the right equipment and helps you in theory and practice in the beginning and aids in teaching you how to use the AERO software package for evaluation, interpretation and automatic reporting.

Process flow for the condition detection of the rotor blades

  • Data collection with consumer drone and data upload to the Aero Enterprise Server

The data collection is done with a consumer drone (e.g., DJI Inspire 2 or Matrice) by you as a drone pilot. Aero Enterprise supports you regarding airborne inspection procedures on the object, process, data handling through training, webinars and documents.

  • Data preparation by Aero Enterprise & Provisioning for you as a partner

The preparation of the data or the post-processing is carried out by Aero Enterprise. These are specific activities such as: exact height determination, positioning in space, image name adjustments, pre-classification of the pictures regarding extra tags like leading edge, trailing edge, rotation angle calculations, allocation of sheet numbers in depending on the rotor position, etc. Then the data for you is customised, secured and ‘ordered’ to be provided to our server. These are encrypted directly via VPN tunnels. The data is therefore not in a cloud.

  • Analysis, qualification and classification of the data by Aero Enterprise or you as a partner

With the AERO-Lysis license software developed by Aero Enterprise you can examine airborne inspection data, mark, measure, document and classify defects. At the end you will get after a short time documentation of the overall condition of the plant ‘by mouse click’ via an entire status report in digital and/or physical form (PDF). This includes the evidence of the damage to the corresponding positions, as well as the footage of the entire plant. The report can feature your branding in the form of logos and sub-information.

Requirements for drone pilots:

  • You need to have a drone licence for the respective country.
  • Drone (type DJI Inspire 2 or Matrice) – flight certificated to use in the respective country.
  • Flight system needs to have an insurance.
  • Ideally, first practical model flight experience and/or familiar with DJI products for more than a year.

Robert Hörmann is CEO/CTO and Founder of Aero Enterprise. His roots are in aviation as an aircraft mechanic, officer and military pilot with the German Air Force. After military service he worked in several companies in Europe in the field of technical sales and business development before he became self-employed in 2013.

Robert Hörmann

Founder & CEO

Aero Enterprise GmbH

+43 7435 21110 100

office@aero-enterprise.com

LEAVE A REPLY

Please enter your comment!
Please enter your name here