Robotics & Autonomous Systems

Find out more about our robotics and autonomous systems testing and validation facilities.

Find out more

Electrical Infrastructure Research Hub

The Catapult has appointed the University of Strathclyde and the University of Manchester to form the Electrical Infrastructure Research Hub.

Find out more

Automation & Engineering Solutions

Find out more about our work in robotics, autonomous systems and artificial intelligence.

Find out more

Stay Current

Dig deeper into the biggest issues facing offshore wind, wave and tidal energy with our series of Analysis & Insight papers.

Find out more

Live Innovation Opportunities

There are a number of programmes laying out the key technology innovation challenges faced by the offshore renewables industry that, when solved, will help drive down the cost of offshore renewable energy, with positive effects for the industry and UK economy. Visit our Live Innovation Opportunities page to find out if your technology has the answer.

Find out more

blog

ORE Catapult Hackathon

Offshore Renewable Energy (ORE) Catapult ran its inaugural Hackathon in October 2019 in collaboration with ScottishPower Renewables and National Grid ESO. The challenge was complex, but the entrants rose to the task, and delivered solutions which exceeded the expectations of both the organisers and the judges.

Wind farms connected to the national grid regularly experience periods of curtailment. This is when they are asked to turn down their electrical output because there is too much power entering the system – wasting electricity and increasing costs. However, when demand increases, there is a lack of electricity being supplied to the grid and as a result, generators must ramp up their supply. Current wind farm capacity is unable to support this.

ORE Catapult Hackathon, October 2019

As the proportion of wind farms connected to the grid rises, the need for them to be able to support this balancing activity will only increase. So why are they unable to do so? Currently National Grid ESO, who operate the national grid, do not have a reliable and accurate measure of the wind energy that could be produced if a wind farm was returned to full operation during a curtailment period.

This electrical energy that could be harnessed is referred to as Power Available (PA). A crude measure is currently available on several modern wind farms, however older infrastructure is unable to meet the strict accuracy requirements.

Our Hackathon teams were asked to develop ways of producing PA signals during curtailment periods using commonly available wind turbine data. This involved developing complex models to create useful, actionable knowledge out of what could be seemingly unconnected datasets.

Around 50 participants over 18 teams gathered in the Glasgow Science Centre to tackle the challenge over the two-day event. It was a tough ask, with many remaining at the venue into the night before returning the next morning to tweak their models and approaches. Midway through the second morning, teams were granted their first access to the test datasets, which could be used to evaluate their solutions.

The models’ outputs were run through an algorithm to evaluate their accuracy, and each model was also compared to ScottishPower’s existing PA data as a sense-check. The top three teams from this numerical assessment and two further judges’ ‘wildcards’ were invited to pitch in the final. Finalists included teams from university students, academics, consultancies and SMEs, with each team taking a unique approach to the challenge, from more traditional data science techniques through to machine learning algorithms and cloud-based analytics environments.

An expert judging panel was brought together with representatives from ORE Catapult, ScottishPower Renewables, National Grid ESO and an independent judge with expertise in the area. All of the teams gave an excellent account of their achievements and methods, however only one team could be the winners.

Offshore Wind Hackathon Winners – Real Wind Turbines Have Power Curves

 

Real Wind Turbines Have Power Curves, made up of current and recently graduated PhD students from the University College London Energy Institute, impressed the judges and claimed first place. The judges were particularly impressed by the team inferring additional wind direction data from the available location and power output data to help create additional parameters for their machine learning algorithm.

Timothy Fletcher, Senior Asset Performance Analysis Engineer at ScottishPower Renewables, said

The team from University College London exceeded our expectations and were thoroughly deserving winners. Despite having limited experience of the wind energy industry, they were able to identify the important features within the data and deploy novel data processing. They tested an impressive range of machine learning techniques within the time available and built a highly effective model to predict power available. In the final assessment they presented their efforts with enthusiasm, clarity and an obvious sense of team work.

Overall, the Hackathon was extremely successful and we hope it will be the first of many. It brought together minds of a very diverse range of backgrounds, and successfully delivered solutions which, with a little development, will be ready to deploy in windfarms across the country.

Alex Louden | Innovation Manager | ORE Catapult

By Alex Louden, Innovation Manager at ORE Catapult

Blog

Related Articles

Cookies on Catapult explained

To comply with EU directives we now provide detailed information about the cookies we use. To find out more about cookies on this site, what they do and how to remove them, see our information about cookies. Click OK to continue using this site.

OK