Today I’m talking to Nicolás Pardo about the project Surecity. It’s a project which aims to give city developers a holistic view of how their future energy mix could look in cities. Nicolás works at the Austrian Institute of Technology and he’s the coordinator for Surecity. His academic background is focused on energy systems, so he seems like the perfect person to coordinate this project. I ask Nicolás to tell me in a single sentence what the project is about. He says, “it’s about creating a platform which allows cities to develop their energy and environment abatement strategies in a easy manner.” Seems like a great aim to me, so I say, “how are you going to do this?”
Nicolás tells me that the project created an energy model in MARKAL/TIMES (a model generator). The results from the model then are visualize for a city’s energy information. These visualisations and the model are available through the project’s online platform. I wonder how the model would apply to a real city. Turns out I don’t have to wonder for long. Nicolás tells me that they are actually testing the model and the platform in three cities: Malmö, Sweden; Judenburg, Austria; and Almada, Portugal. I ask Nicolás, “why these cities?”
Nicolás explains that the three cities each represent three completely different European climates: Malmö is in northern Europe which tends to get colder in the winter, Almada represents a warmer coastal region, and Judenburg a more temperate climate. Between them, they basically cover all the different kinds of climate European cities tend to be in. Additionally, the three cities are all different sizes. These factors help the project to take a generic and wide-ranging approach; ideally the model should be applicable to all European cities. I ask, “how can you effectively run such a large-scale project across three cities in three completely different countries?” Nicolás excitedly tells me that the project has several partners who really make this possible.
The project partners in Austria are the city of Judenburg, the Austrian Institute of Technology, and the Styrian Energy Agency. In Sweden the partners are the IVL Swedish Environmental Research Institute, Luleå University of Technology, and the city of Malmö. Finally, in Portugal the partners are the municipality of Almada and the company called 3Drivers supported by Cneter for Enviromental and sustainability research of at Universidade Nova de Lisboa. I think to myself, that’s a lot of partners but I guess it’s necessary for such a complex project. I’m sure Nicolás has a serious task on his hand coordinating so many partners. He says to me, “the project is definitely challenging.”
Nicolás explains the project faces all of its challenges systematically. Firstly, the project identifies what the needs and requirements of a city are. This means that the project has to pay particular attention to which kind of indicators matter to the city. Then, the project has to consider the needs of the commercial, residential, and municipal sectors within a city. However, that isn’t it; following this the project has to account for public lighting, waste water management, waste management, transport infrastructure, the electricity structures within a city, and the types of energy supply the city has to name but a few factors. I think to myself that’s a lot of things to account for.
Finding this all a bit overwhelming, I deice it’s best to approach this all from a different perspective, I ask Nicolás “how will the platform work from the perspective of a user?” Nicolás explains that a user can follow a link from the platform’s website which will allow them to download the necessary software to connect to the Cloud where the platform is located, then they can upload their data (population data, buildings data, energy consumption, etc.) to the platform. After this, the information inputted automatically generates a model. This platform then generates visualisations for the user from the model results. The visualisation not only holistically show the energy distribution in a city but also produces predictions for how these energy flows and consumption patterns will develop in the future. Nicolás says, “basically you can see maps and energy flows which show energy consumption by sector in each area of the city. Not only that you can see nice clear diagrams which show future energy consumption patterns.” I think this all sounds great; it must be super handy being able to see all this information in a holistic and simple way. Nicolás is careful to remind me that it’s not just about seeing how things are at the moment but also about making accurate forecasts for the future. In fact, the model predicts energy changes up to 2050. I think it’s pretty amazing to able to see the way a city could develop thirty years from now!
I begin to reflect on how the model calculates it predictions. Nicolás tells me that like all models there are a lot of calculations built into it. The model is of course dependent on the data which is provided. Nevertheless, the model aims to be a heuristic which can be applied to many cities. Nicolás emphasises that a lot can be worked out by the basic information cities already possess, they simply don’t know how to use it. This is where the expertise of the project comes in because the model does know what to do with this information. Nicolás tells me that they approached the project with two levels of data input: modelling based on basic information which most cities possess, and modelling based on specific information from the cities in the project.
I tell Nicolás, I’m really curious about how the prediction element works. He explains that to understand the model’s predictive power we have to consider its parts: MARKAL/ TIMES was created by the ETSAP group supported by International Energy Agency for the very purpose of seeing energy flows and decisions in the context of optimising costs. So, in that regard it does this incredibly accurately. The innovative aspect of the project was to use MARKAL/ TIMES to create a model from scratch which generates complex simulations showing how energy flow would develop at city level. In particular, the model factors in sustainability goals. For example, if your city wants to reduce C02 emissions by 80% by 2050, then the model will factor this in when generating results in the form of visualisations. Crucially, the project focuses on making sure the model is well connected to the online platform.
Nicolás makes this all a bit less abstract by pointing to how the project painstakingly considered every goal a city has. For example, in Almada they knew the municipality was focused on the use of cooling systems whilst in Malmö the municipality wanted to focus on heat pumps. Nicolás points out that size doesn’t matter much as the cities’ objectives are very similar. This prompts me into asking, “if not objectives, what was different between the cities?”
Nicolás points to the communication system built into the platform. For Malmö this was less interesting because it was a city large enough to have its own communication team. Whilst Judenberg is a city which is small enough to inform its citizens through social media. However, for Almada, the mid-sized city, the communication aspect of the platform were more interested as it was too small to have a communications team and too big to reach citizens effectively by social media alone.
Now that I have a reasonable idea of how the platform works I press Nicolás into telling me who benefits the most from the platform. He says, “the answer isn’t so simple, actually a lot of people benefit.” He explains that municipalities benefit the most as they can better allocate their resources using the platform. Having said that, citizens also benefit because through the platform they not only get to see how energy is used in their city but also how the municipality plans to develop energy strategy and how it will achieve their sustainability goals. Finally, businesses also benefit because they can see where the areas of growth are. For example, a company selling solar or heat pumps might see that in a particular city there will be a large market for them in the future, so they can then meet this need. In the end, the project develops a common understanding of the city in the eyes of the municipality, citizens and business.
Nicolás wants to stress the fact that model really can adapt to your goals. Let’s say your city wants to decarbonise by twenty percent in five years, the model will factor this in when offering a prediction. However, if you want to change the decarbonisation percentage the model can easily adjust its prediction. I’m pretty impressed so far but I still feel like I haven’t addressed an important question, “how do you know the predictions are reliable?”
Nicolás says that this is of course the most important thing. He tells me that we have to consider the constraints put on a model, such as the future number of insulated houses in the city. These constraints are what determine the reliability of the model. He explains this further by saying that when adding inputs into the model we have to be sure these constraints fit with reality. Yet, Nicolás wants to draw attention to the fact that perhaps one input can be inaccurate, but it is practically impossible for the majority of inputs to be wrong. He says, “we can’t predict the future exactly, but we can accurately predict trends.”
We can’t predict the future exactly, but we can accurately predict trends.
Feeling relatively satisfied that I understand how the model creates accurate predictions, I decide it’s time to focus on the other interesting thing about the platform. I ask Nicolás about what he said earlier regarding the project producing holistic views. He seems keen to talk about it, “the model shows how different constraints interact with each other and affect the rest of the system.” I ask him for an example. He points to the case of insulation and district heating: they tend to compete against each other. Nicolás tells me, “you typically either argue to increase insulation or argue to increase heating, not both. However, with the model you can actually see the optimal balance because you can see how the whole system reacts to a change in terms of energy flow and even marginal prices”
I tell Nicolás that the methodology must be really complex, he agrees with me. Yet, he tells me that from the cities’ perspective they’ve made it really simple to use the platform. They did this by giving a lot of care and attention to what users need. Nicolás says, “we had to carefully define what aspects the cities want to see in terms of technology, geographical level and temporal units.” All this consideration went into the making of the model and the platform. He adds, “the average person using this platform won’t be an engineer or a data scientist and we wanted to make sure they could use it.”
I ask Nicolás if the project encountered any unexpected difficulties. He laughs, “yes a lot, but we’ve resolved them!” I think it’s a good idea to focus on the biggest one as we’re running out of time. Nicolás tells me that the sheer size of the model (due to the inputs and calculations) means it was difficult to have everything represented on a single platform in a comprehensible manner. He assures me that in the end they managed to do just that. This automatically prompts me to ask, “are the results ready to be scaled up?”
Nicolás responds emphatically, “yes, our model can be applied to any city, as I said earlier we’ve found out that the size of a city really doesn’t matter and we’ve already tested it in three cities with completely different climates.” I feel pretty impressed that they’ve managed to achieve this within a single project.
I finish the interview by asking, “what was the most interesting and unexpected discovery within the project?” Nicolás laughingly answers, “most people don’t care how we do our calculations as long as they are correct. They just want something clear and comprehensible to use. I guess the most unexpected thing was that in terms of data usage; most cities didn’t know what they have and they didn’t know what to do, but we’ve thankfully got a solution for them.”