We transform data about electric power into information, says the CEO of Mycroft Mind

Filip Procházka

„I used to work on simulations of life in an anthill, but they told me that there was no way I would get paid for that.“

Your work is not easy to understand for an outsider. In essence, you take a huge amount of data available on electric energy flows and you analyse it. Is that right?

That’s partially right, but the catch is that the data is often not really available. Distribution networks are increasingly impacted by renewable energy, which influences how the networks work, and this trend is only getting stronger. That’s why we need better and more detailed data for network management. The CEZ Group, for instance, has 100,000 kilometres of low-voltage power transmission lines, three and a half million customers, and about 50,000 electrical substations. There's still not much data collected at the low-voltage level. We develop models and simulations, collect data, and try to transform this data into information to help inform the operators’ decisions and improve the distribution network management.

As an IT expert, did you just think that this would be a good thing to do or did you get this request from the energy companies?

It’s a global trend called smart grids. The traditional top-down management is gradually moving to lower and lower levels. Instead of hundreds of sensors, we now have tens of thousands, and in the case of smart grids, there are millions of them. You cannot analyse data like this using standard IT methods. The energy companies currently work with one number per customer, and that is the total yearly consumption. However, this number alone is becoming increasingly insufficient. They need more detailed data not only about consumption but also about voltage and any problems detected in the network. Once they get this data, they get into a state that’s now called data shock. You need highly specialised experts to analyse such volumes of data and it takes a lot of research. This means that the energy companies aren’t just looking for someone who will sell them a really neat software solution; what they need is a partner for this data adventure. You have to work together to get the results you want.

What are the biggest problems currently facing the energy companies you work with?

Currently, their biggest concern is how to handle the onset of new technologies: whether to do as little as possible to meet the minimum requirements set by the law, or whether to take a more strategic approach and seize the opportunity to take more profound steps in energy distribution networks development.

And are all three companies reaching the same conclusion?

All of them view the situation in a similar way, the differences between them stem from the differences in the areas they serve. Pražská energetika operates in Prague, which means that their network is spread over a smaller area. There are many more underground cables and fewer overhead power lines, and the reliability requirements are higher. CEZ and E.ON, on the other hand, operate over larger areas that include cities, towns, villages, fields, and forests, which means that communication with facilities in the field is much more demanding. However, it’s really nice to see that all three companies take this seriously, even though it was tough at the beginning.

What was it like, then?

When Mycroft was starting out, we looked at the problem from a purely IT point of view. Our assumption was that while the amount of data keeps getting larger, our ability to understand it does not improve that quickly, so there was a space for a company like ours. You might think that our only problem would be to pick and choose who to work with. And you would be wrong. Because it’s one thing to explain the concept in theory like this, but it’s a completely different thing for a customer to realise that they actually need this. To help our customers, we have to understand their data, but to understand their data, we need to have some experience with it first. That’s a vicious circle for you. We took many wrong turns. One of the projects we worked on was processing petrol station data and identifying fraud by the station attendants or tanker drivers. However, we often hit a wall when the company, even though it was aware of the problem, was part of a global corporation and didn’t have the authority to deal with it in the Czech Republic. And sometimes the solution we developed was like using a sledgehammer to crack a nut. But as I like to say, the problems in energy are a worthy opponent. You really need a sledgehammer to crack them – a gavel won’t get you anywhere.

When you enrolled at the Faculty of Informatics sometime in 1998/1999, even the internet was far from ubiquitous. How did you imagine your future job back then?

For a number of years, I had worked on a system that eventually turned into our modelling and simulation tool. Back then, I worked on a simulation environment for creating computer game characters. I was no gamer, but I wanted to do something to make the worlds created by games more interesting. That’s why I enrolled in IT. I really enjoyed this and I went so far as to simulate life in an anthill but, as Associate Professor Račanský explained to me, there was no way I would get paid for that. Then I met Zdenko Staníček; we hit it off right from the start and started doing what I’m still doing now for a living.

You also completed your PhD at the faculty before leaving to build the company, but when the university started the science and technology park, you were one of the first to move your company here. What do you expect from this?

I see it as an opportunity to be in an inspirational environment. Translating science into practice takes a long time, but the benefits are real, so you just have to persist. And as I already said, the problems in energy are really worth the effort as it’s one of the most crucial infrastructures there is. Any blackout is going to have very quick and drastic consequences. There are many interesting things in this field that are worth getting stuck in to and it would be a shame if people around the university didn’t get involved.

In two years, your company will celebrate its 10th anniversary. Do you have plans for where you want to take your business in the future?

I don’t really see myself as a businessman. I would rather say I’m an architect. Some software developers chase big US companies so that they can put their impressive names in their CVs. However, you have to decide whether you want to implement a new accounting system thirty times in your career, or whether you want to develop software for space shuttles. And I’m all for the space shuttles. I don’t have specific goals for how much money we are going to make or when and how to make it. The way I see it is that if we do it well, it will be profitable. But you have to start with doing it well.

Alumni Gallery