Meet our Energizer! Pedro Vergara Barrios
Pedro Vergara Barrios is from Colombia and obtained his PhD in Electrical Engineering from the University of Campinas, UNICAMP, Brazil, and the University of Southern Denmark, SDU, Denmark in a double-degree program. Four years ago, Pedro moved to the Netherlands. He worked in a post doc position in the Electrical Energy Systems group at the Eindhoven University of Technology, before taking a position as an Assistant Professor in the Intelligent Electrical Power Grids group at the Delft University of Technology.
Pedro's research topics are related to the development of algorithms and methods for control, operation and planning of electrical distribution networks with high penetration of low-carbon energy resources (for example, electrical vehicles, PV systems, electric heat pumps) using classical optimization and machine learning models.
Energy transition close to home
In our day to day lives, we all aim to use electricity at home, for cooking, the laundry, etc., since it’s a more sustainable energy source compared to gas. When you think about it, you could ask yourself why aren’t we part of an interactive supply and demand network using only renewable energy yet? Let’s take a closer look at current developments in electrical distribution networks.
Pedro’s research is in developing algorithms for planning and operation of electrical distribution networks, at a city level. This level is somewhat different compared to the infrastructure employed at a national level. Pedro explains: “Both networks are connected, but if we go down to the city level, the voltage level, the modelling assumptions and the physics are different compared to the electricity infrastructure on a national level. Therefore, one person focuses on this part, and that’s me.” The research into planning and operation of electrical distribution networks is directly aimed at supporting system operators, such as Liander; Enexis and Stedin, they work within technical limits for distributing resources in the network and they secure the energy supply all year round.
Accommodating as many renewable resources on the distribution network as possible
One way of contributing to the energy transition, is by making sure that we can accommodate as much as possible low-carbon (re)sources, such as solar panels; electrical vehicles (as resources for generation and consumption of electricity); heat pumps; etc.; in the city-level electrical distribution network. This is a complex task with a number of challenges.
The network should be able to distribute these resources to households, observing all technical requirements, securing a steady energy supply. In this new situation, the different requirements for maintenance and operation of the networks call for data-driven solutions. By using sensors, system operators are able to collect more data and get more insights in the supply and consumption of renewable resources in the distribution networks.
“Developing algorithms based on machine learning seems to me the obvious starting point for developing data-driven solutions. System operators are installing more sensors and collect more data. Customers generate more data by using smart devices, such as electrical vehicles that have their own apps and collect data on energy consumption. In my research, I try to make use of all these available data to improve and coordinate the use of these low-carbon energy resources, with the aim of accommodating for as much renewable resources on the distribution network as possible.”
Working at the interception of electrical engineering and machine learning
When Pedro explains his research to interested parties, he often finds that, in general, we make the assumption that people would like to share energy data. However, this is not the case, although people want to contribute to the energy transition, energy data can carry private and personal information, thus, getting access to such data is an obstacle.
The approach to deal with these obstacles, according to Pedro, is to make use of open source data. “The system operators release data with consent of customers. However, before we are able to accessing these data, the data on hand gets a bit out dated. Therefore, we are looking into using Artificial Intelligence. Using AI might help in extrapolating these datasets and design several assumptions in the models for better planning. This actually something that really has my interest, to be able to work at the interception of electrical engineering and machine learning.”
Invisible congestions and long-term solutions
Another difficulty that Pedro comes across is explaining that distribution networks have their limits. Limiting congestions on a high way are very visible and tangible. By contrast, congestions on the electricity network are somehow invisible and people do not tend to consider the possibility of limitations to the distribution networks. This can lead to assumptions and demanding questions to install more solar panels, to install more electricity heating systems, etc. However, in such cases, the network operators have to tell their customers ‘no’, and that’s sometimes an answer difficult to accept.
Pedro elaborates: “Take for example the installation and use of electric heat pumps: there is no capacity for this at the moment. The high price of gas is putting pressure on the discussion to renew or adapt infrastructure. However, the infrastructure is costly and takes many years to build. The obvious solution is to invest in infrastructure, but then we are talking about the long term. For one neighbourhood it can take five to seven years to renew the energy distribution network. It takes even more time when working in a historical neighbourhood, such as in Delft or in Amsterdam.”
Let’s look at our own energy consumption
So, not all solutions are easy to implement, we have a long way ahead of us before we are part of an interactive supply and demand network using only renewable energy. What can we do in the meantime? Pedro: “In my personal opinion of the current developments in the energy transition, I would say that we need to participate all together in the energy transition, not only the government, not only operators, but also you and I, all of us. Perhaps we can change in our energy consumption and use energy when it’s available. It’s not only up to the owners of the infrastructure.”
Small-scale 4TU.Energy workshops make a huge difference
Asked about 4TU.Energy, Pedro suggests that 4TU.Energy can contribute to the energy transition by bringing together different expertise on energy. “It was great to attend a small workshop in Twente earlier this year, where I met people from Electrical Engineering and Computer Science. We don’t need to see each other every day, but it’s good to meet every once in a while.”
Pedro is Assistant Professor in the Intelligent Electrical Power Grids group at the Delft University of Technology. If you would like to know more, please contact him via his personal webpage.