Wednesday, 28 September 2016
14:30 - 16:00 Droning On! The use of UAVs in wind turbine O&M
O&M & logistics  
          

This session will give an overview of the emerging technology of unmanned aerial vehicles (UAVs), otherwise called drones, and their application to the remote inspection of wind turbines, with an emphasis on the assessment of blade damage. It will look at what has been done to date and what the capabilities may be for the future. Finally, it will consider the advantages and disadvantages of different inspection techniques.

You attended this session?

Please give us feedback

 

Learning objectives

• List the ways in which wind turbine blade damage can be assessed using drones;
• Identify the potential advantages of drones over other approaches to turbine assessment;
• Anticipate future developments in drone technology and associated instruments.

Co-chair(s):

Lars Landberg DNV GL Energy, Denmark
Co-authors:
Lars Landberg (1) ^F Elizabeth Traiger (1) Tom Richardson (2) Chris Day (3)
(1) DNV GL, Hellerup, Denmark (2) Bristol University, Bristol, United Kingdom (3) Schiebel, Vienna, Austria

The use of smart drones in wind energy

Introduction

This presentation will focus on the use and possibilities of Smart Drones for inspections. The technology is in many areas in its infancy, but there are clear indications where things will go, these will be described in this presentation and examples given.

Approach

The approach will be that the State-of-the-art will be described and then the technology and applications of smart drones discussed at length.

Main body of abstract

Drones are currently being used in the wind energy industry at an “entry level”, but there is no real consensus on UAV applications. DNV GL sees many players, but we do not yet see standard ways of using drones and full utilization the technological capabilities and detailed information obtained. The first part of this presentation will give an introduction to the state of the art. Here we will touch on typical characteristics of present day drones, the current regulations and give some examples of drone-derived footage of wind turbine inspections. The main part of the presentation will be about smart drones.

Present day commercial drones have a number of shortcomings including minimal automation, limited redundancy in the vehicle design and relatively high requirements for operator experience and knowledge. The effect of these has been to limit the impact on application, efficiency and cost.

The focus of the presentation will be on ‘Smart drones’ which will be highly automated, carry a wide range of sensors and which will significantly reduce the current operating risks and costs whilst significantly increasing the capability.

Techniques that offer significant potential to drive this evolution in aerial intelligence include:

1. Combinations of techniques including vision based navigation and sensor fusion to remove the requirement for piloting skills and to optimise imagery collection.
2. Adapting manned commercial aircraft avionics architectures with MEMS technology to deliver dramatic improvements in reliability and safety.
3. Novel configurations and control system design, providing enhanced tolerance to winds and gusts.
4. Multispectral imagery combined with digital analytics derived from the security/defence industry will provide robust high quality, low false alarm, automated damage detection.

System autonomy and precision control, provide opportunities to optimise flight times and drive inspection times to a minimum, these savings can then be used to drive down inspection costs or increase the frequency of data collection.

In parallel to the platform development there will be an expansion of operations in terms of the allowable flight envelope and remote operation, up to and including BVLoS (Beyond Visual Line of Sight). With end-to-end single-button operation the capability will be developed to allow one person to operate more than one smart drone safely – thereby increasing efficiency and reducing costs.

An important aspect of the effective and efficient use of drones is the use of AI in “looking at” and analysing the output. Large volumes of real-time high-definition video and other sensor output is generated from each flight. In the presentation we will give examples of how existing image processing techniques (from the medical field) supplemented with deep learning systems that can analyse the footage, frame by frame, along with external sensor data to identify issues that would need further (human) investigation

Conclusion

This presentation has given an overview of the current use of drones and future use of smart drones. The use of drones and especially smart drones will make iinspections of wind turbines more effective and efficeint. Elements that have been discussed include: autonomous flight, AI and deep learning for image processing, and multi sensor utilisation.


Learning objectives
* A good understanding of where drones and drone usage in wind energy are today, this will include limitations on both technical and regulatory highlighting where innovation is possible.

* A view into the future of the potential new technologies and techniques that can be used to develop Smart Drones and significantly expand their use and application.