A powerful UAV-based ICT solution allowing safe, reliable and effective inspections of wind turbines

Informations projet

Wind-Drone

N° de convention de subvention: 782517

DOI

10.3030/782517

Projet clôturé

Date de signature de la CE 1 Août 2017

Date de début 1 Septembre 2017

Date de fin 31 Decembre 2017

Financé au titre de

• SOCIETAL CHALLENGES - Secure, clean and efficient energy

Coût total

€ 71 429,00

Contribution de l’UE

€ 50 000,00

Coordonné par

BLADEINSIGHT SA
Portugal

Periodic Reporting for period 1 - Wind-Drone (A powerful UAV-based ICT solution allowing safe, reliable and effective inspections of wind turbines)

Période du rapport: 2017-09-01 au 2017-12-31

Résumé du contexte et des objectifs généraux du projet

The inspection process in wind turbine blades still relies, in its large majority, on qualified inspectors roping down each blade. This process has an average cost of €1500 per turbine due to the skilled labour involved (includes risk compensation), expensive equipment and machinery rental, as well as exceedingly costly transportation for offshore inspections.

Our WindDrone technology is a turn-key solution to revolutionise blade inspections of wind turbines. The customised, off-the-shelf UAV is equipped with proprietary control software and highly accurate inspection and imaging equipment. The drone autonomously follows an optimal path that covers the entire blade, all the while enforcing a minimal safety distance from the blades, thus eliminating the risk of collisions. Once the images of the blade are taken and the flight is over, the inspection data is then sent to our user-friendly cloud software platform, BladeInsight. The data is stored and autonomously processed to enable a powerful analysis of faults through deep learning algorithms. WindDrone’s centralised database collects and learns from the status of every blade and the location of damages. The faults are then automatically uploaded to a user accessible dashboard in the form of personalised reports. These reports provide a list of all damages identified, their picture as well as classification of their severity and guidelines for intervention and planning of concise and targeted repair operations.

In 2014, the European Union set a legally binding target by 2030 of at least 27% of final energy consumption in Europe to come from renewable energy sources. Wind energy is poised to take the lion’s share of renewable energy demand, with 21% electrical energy generation. The prospective 96.000 wind turbines planned for installation on land and offshore, could avoid emissions of up to 436 million of tonnes (Mt) of CO2 annually. The only realistic approach however, to increase adoption on a wider scale, is to reduce what we call Levelised Cost of Energy (LCOE) to be competitive with current-day fossil fuel providers. Multiple factors add onto LCOE, including cost of installation, operations, as well as maintenance of renewable energy generator mediums. In the case of renewable wind energy, apart from capital outlay, operations and maintenance costs are core determinants of LCOE . In addition, the traditional wind turbine inspection process has been cited as hazardous with great need for precaution . Blade rope work is a risk-laden and expensive activity, which should only be kept where necessary; in targeted repair work. As such, further efforts must be taken to consolidate and improve Occupational Safety and Health (OSH).

Travail effectué depuis le début du projet jusqu’à la fin de la période considérée dans le rapport et principaux résultats atteints jusqu’à présent

This study proved instrumental to investigating the market landscape, strengths, as well as meeting requirements, to ensure successful first revenues from the proprietary technology. The Feasibility Study showed that a new business model was needed to approach the market and achieve higher rates of market penetration. As such, first commercial entry has been identified through the interest of wind turbine inspection companies. The main achievements reported in the study were:
• Upon further in-depth market analysis through investigations and meetings with key market players, we found that the problems and needs in this market aren’t being adequately met by the current blade inspection methods.
• The market demand was validated through discussions with current customers and end-beneficiaries. This showed that the wind inspection companies and wind farm operators desired faster, safer, and more cost-effective inspections.
• As a result of the above, we found that the most beneficial business model to approach the market would not be to sell the solution to wind farm operators but instead to the wind turbine inspection companies. The reasons behind this included:
o Wind turbine inspection companies already have a large network of customers and clients in many different European countries, as well as other geographical markets we will be targeting in a few years after commercialisation;
o The lower amount of customers will mean less UAVs are required for inspections. As a result, no automated assembly process is need due to the lower numbers of UAVs to be produced;
o Inspection companies operate around the world, as well as operating in other inspection markets that could prove very beneficial for Pro-Drone if we look to expand our solution to other vertical structures and inaccessible systems, such as bridges, chimneys, offshore rigs, etc. This would give us an easy market entry point, with an already secured customer for a quick market takeover.
• The wind energy and renewable technology markets’ current trends show that the increasing growth of renewable technology will continue as many developed countries aim to reach their target goals for 2020, 2030, and 2050. In this current climate, wind turbines are proving to be the most easily implemented and installed technology and as such is leading the front for renewable technologies. This means that the targeted market will continue to grow in the future and offer an opportunity at a global market.
• The MVP of the WindDrone solution was finalised through stakeholder feedback. This resulted in the use of off-the-shelf, and easily accessible drones and inspection components as part of the MVP. Instead of the UAV being the main part of the solution, the software algorithms are. The UAV’s flight control, collision avoidance, and imaging algorithms are combined with the cloud platform’s fault detection, report generating, and machine learning algorithms to create the overall WindDrone solution.

Progrès au-delà de l’état des connaissances et impact potentiel prévu (y compris l’impact socio-économique et les conséquences sociétales plus larges du projet jusqu’à présent)

The WindDrone solution is well positioned in the market in recent years due to a number of different reasons. The first being that drones are becoming much more widely used in a number of different industries, with them becoming extremely commonplace in businesses . Delivery companies and retailers, such as UPS and Amazon, were the early adopters of drones in businesses, but are present in many different sectors like media, agriculture, and industrial inspection to name a few. The constant advancements in UAV technology have created cost-effective, autonomous drones. Due to this rapid progression in UAV technology and a large increase in adoption of UAVs, it has become a perfect time to enter the market with a solution capable of automatically analysing vertical structures like wind turbines.

In the last few decades, a focus has been placed on getting energy and electricity from renewable energy sources, resulting in many organisations and institutions researching renewable technology and creating possible clean sources of electricity. Wind power was one of the first developed and is currently the second most widely used behind hydropower. With 486.8 GW of wind energy installed worldwide, and a CAGR of over 11% (55.6 GW) in 2016, wind energy is the fastest growing renewable energy technology at the moment. Europe follow Asia as the second largest producer of wind energy. For this reason, the wind energy market proves to be one of the best market opportunities to enter with this solution as expectations have the total number of wind turbines worldwide increasing from ~350K in 2016 up to over 600K in 2021.

Much of the wind turbines installed over the past decades have begun to age and wear. This results in small faults such as cracks and imperfections to appear and propagate on the wind turbine’s blades. A turbine’s energy output can reduce by as much as 12% over the course of its 20 year lifetime. As such, maintenance of these turbines is extremely important and can account for up to 25% of the total levelised cost per kWh produced in their lifetime. These problems with the ageing renewable infrastructures have caused the market to need a solution that can help improve the current, expensive maintenance methods.

Finally, the recent developments in machine learning technology have helped pave a way for the solution in the current market. We utilise this technology to automatically detect any faults in the blades through collection of various data points from multiple in-field tests with partner wind farm operators. This is extremely beneficial as it eliminates the need, and cost, of an engineer to manually go through each image and detect each flaw. This information would then need to be written up in a report, which is clearly very time consuming and not possible on a large scale. Hence with the advancement of machine learning, the solution can improve inspections and introduce a new method of processing the wind turbine blade analysis much more efficiently and progress to a global scale.

pro-drone.png