Fortum has a vision about a cleaner world, by reshaping the energy system, improving resource efficiency and providing smart solutions. Data-rich, AI-fueled solutions play an important role in the realization of this vision, according to CDO Per Edoff.
The energy system is changing. In the old system, with its centralized production, the challenge was to distribute the resources. Now, with decentralized resources like solar and wind, customers become producers and the challenge is instead about optimizing the systems. According to Per Edoff, CDO at Fortum, we now need an increased flexibility both in power production and in consumption.
Here, emerging technologies such as IoT and AI are important tools to capture, make sense of and act on data. During the Snapshot event AI for Good, Per Edoff shared tree examples from Fortum about how AI supports the optimization and flexibility to make the energy system more efficient:
- AI-optimized district heating. The temperature has been an input to the production system, but previously there was no feedback loop from the consumption side as such. Now, with the help of AI, Fortum knows that there is a slowness in the houses, and if the forecast for example tells about a limited drop in temperature, they don’t have to start peak production (which is sometimes fossil based). Instead, Fortum can utilize the capacity that is already in the systems, something that will save fuel and lower emissions. As more wind and solar power comes into the system, an algorithm also helps to optimize the heat day by day and to store heat produced by cheap electricity.
- AI-powered electricity site. The same logic – to avoid peak production – is applied at the electricity site. In our houses, we have a lot of always-on energy consuming functionalities without thinking about it. Like the water heating boiler; constantly on but only used for shorter periods of time. If there is a problem in the network balance, the AI-based solution can now shut down the heating system which means that the grid owner doesn’t have to order peak production. And the customers won’t notice anything.
- Smart Living project. The SmartLiving pilot project in Stockholm involves the consumers and addresses energy management. Here, energy consumption is monitored and visualized in real time: water, heat and electricity of the whole building, but also in each apartment. This makes it possible to, for example, choose to do the laundry off peak time when the impact is low.
What about the future then? Per Edoff says that the winner in the utilities sector will be the one who can optimize the whole system. With an increase in solar panels and alternative energy sources, there won’t be any lack of energy. Rather, the capacity, within an underestimated network, will be the real bottleneck. An increased flexibility on both the production and the consumer side, and control over the data with strengthened capabilities to process information with AI, will be necessary.
If you'd like to know more, check out the whole AI for Good seminar which is still available ondemand.
AI: Ready for Business
Within most organizations, there’s typically a single business issue or goal that kicks off the initial AI journey. After that, its development rarely follows a straight line.