Embracing the Data Science Challenge: TNEI's new Data Science team

Data Science
10 Mar 2021

TNEI is proud to officially launch our Data Science team! Over the past several years, our dedicated data scientists have been busy building up our capabilities and exploring the different ways we can support our core power systems services with more data-driven expertise and analysis. Now, with a clear direction of travel within the energy sector towards the adoption of highly sophisticated analytical methods and model development, the time has come to officially launch the team. 

Data Science is Becoming Essential

Data is currently a hot topic in the energy sector. The joint BEIS, Ofgem and InnovateUK programme on Modernising Energy Data, which is leading activities to develop best practice and promote visibility of data, is a great example of the level of activity going on in the area, and signifies the strategic importance of data by policy makers and Government. Ofgem is also reflecting these ambitions within the next round of RIIO price controls. Several network operators, including National Grid ESO, have started to provide open data portals on their websites.

However, gathering data and making it accessible is only the first step. As a sector, we also need to modernise how we use that data once it has been made available. Analysts and decision-makers need to address what we can do with the data that has been gathered, and how can it help the sector face our very significant challenges as we work towards Net-Zero. The only way forward, we believe, is to embrace data science – the art and science of using data sets to gain insight, predict and control.

As a result, TNEI has established our new Data Science team to help address these challenges and integrate data science into the energy sector.

Stephanie Hay, Director of Networks and Innovation, comments: "Our power system is evolving drastically, and there is a lot of momentum behind this. To plan, operate and manage networks will require thinking and capabilities outside of what we would consider traditional electrical engineering. Recognising this, TNEI took some tentative steps into the world of data science a few years ago. Since then, the team has grown rapidly, and delivered some truly ground-breaking work for our clients. TNEI believes Data Science is critical to the advancement of the power system and we are delighted to have a strong team of experts leading the way."

Why TNEI?

In some ways, it is easier than ever to have a go at doing Data Science. Download a few open-source software packages, load in some datasets, and go.

So, what can TNEI offer?

Most importantly, all our data scientists are, first and foremost, energy sector experts. One of the key skills for a data scientist is domain expertise and we have decades of experience in our team, covering issues like control room operations, long-term planning, network economics etc. This domain expertise means we really know what problems we are trying to solve, we understand what sort of decisions need to be made and we won’t get easily misled by any spurious results!

In addition to being energy sector experts, we also have a very high level of statistical expertise, with several team members joining TNEI after working in academic research. This combination of domain expertise and statistical proficiency means we have a different perspective on data science and machine learning to many others. We are generally less interested in off-the-shelf tools and black box models – although we appreciate that they have their place – rather being more interested in bespoke tools and explainable models, that can aid transparent and optimal decision making. In such a heavily regulated industry,  this is important. Plus, with the rollout of technologies such as smart meters and electric vehicles being very gradual, the problems faced by the industry often involve small data sets, rather than big data, and this tends to allow maximum use of our statistical modelling skills.

We’ve already delivered many data science projects, looking at topics such as making predictions and reinforcement decisions aided by sparse smart meter datasets and understanding the volatility in wind farm output in relation to providing nearly firm ancillary services. Current projects include:

  • long-term network planning under uncertainty for complex technical constraints
  • short term probabilistic forecasting for control room operations
  • simulating the highly variable and uncertain interactions of various flexibility markets
  • calculating the impact on consumers of unusually frequent faults.

Since extensive use of data science methods is quite novel within the energy sector, most (but certainly not all) of our work to date has been on innovation projects. Our Data Science Team is therefore currently embedded within TNEI’s Networks and Innovation team, but this hasn’t stopped us from looking at wider applications, such as sophisticated predictions of wind farm noise phenomena with our Environment & Engineering colleagues.

Our data science work is led by Dr Gruffudd Edwards (Senior Data Scientist) and Gordon McFadzean (Principal Consultant). Gruff has considerable experience in mining, analysing, interpreting, predicting, and visualising data. His academic background spans statistics, electrical engineering and physics and his research has often bridged these disciplines, including a PhD on advanced time series modelling of wind farm power availability.

Gordon has a broad background which includes energy, engineering, and economics, with a track record in delivering innovation projects in the power systems sector. He has experience of technical issues related to network planning and operation as well network charging. His current focus areas are LV networks, network flexibility and charging reform.

The most recent addition to the Data Science team is Sarah Sheehy, who has made the transition from academic research to join us, adding to the team’s expertise in statistical modelling and analysis of power systems. Her academic background includes engineering, statistics and operational research, and her research has combined elements of each discipline to investigate a variety of power and energy systems problems. Sarah has also spent time working in policy, undertaking a secondment to the Cabinet Office to work on academic collaborations with policy makers for data science projects within government.

How can TNEI help me embrace the data science revolution?

TNEI has expertise in and can assist with the following services:

  • Forecasting,
  • Machine learning,          
  • Data visualisation and interactive dashboards,   
  • Optimisation and decision-support, and
  • Probabilistic modelling and statistical simulation.

View more on our data science pages

Resources

Our data scientists have already published several articles that cover some of the most relevant data science topics for our sector, including:

We have also hosted webinars on the application of data science in the energy sector as part of our TNEI Insights series. Join our Forecasting TNEI insights webinar: Data Science in the Energy Sector - The Forecast is Bright on Thursday 29th April 2021 where we will build on our recent Forecasting 101 article.

We hope you will fully embrace the opportunities offered by the data revolution and to let TNEI be your trusted guide to the cutting edge of data science.

View our data science services to find out how your project can benefit from this way of thinking or get in touch with our experts.