2 years ago
The energy market strained to the limit - AI creates a more sustainable and efficient future

The energy market is becoming increasingly complex. Completely new types of energy are added continuously and the proportion of variable renewable energy (VRE) is growing rapidly. For energy companies, the use of AI has become a matter of survival.

The European energy market is becoming increasingly complex. The number of different renewable energy sources is growing continuously. The number of smart buildings and vehicles for electricity networks (V2G) is increasing, at the same time as the energy markets are becoming more regional and fast-moving. And on the consumer side, new platforms are constantly being developed, which will help people to plan, buy and sell their electricity production and use based on local and personal conditions.

For the energy industry, this means a completely new, unmanageable reality, where factors such as start-up times for production, market balancing, short-term storage and transmission effects are becoming very important.

- The development is starting to become so rapid and granular that it is no longer a matter of months, weeks, days or hours - all optimization must now take place on a quarterly basis and preferably every microsecond of the day in an energy market that is now European and no longer national. This year alone, we have learned how German gas prices affect households in Sweden, these are connections that did not previously exist, says Andreas Wiklander, Head of Operation at Jämtkraft.

A growing concern in the market

Since the new, green energy sources are often less predictable than fossil fuels and nuclear power, it is no longer primarily the amount of energy that determines the economic value of an energy source. The exact time of production and the ability to predict this time becomes much more important. Increasing shares of VRE in the market lead to a higher share of losses for each producer due to the weather-related correlation in production between producers.

- Network stability becomes a problem on a much shorter time scale in the form of microseconds to seconds. New, green energy sources are simply less predictable than fossil fuels and nuclear power. Therefore, there is a growing concern in the market for increased inefficiency and system errors, says Andreas Wiklander.

Weather fluctuations can create significant differences and effectively divide the markets into several separate pricing areas, where an equilibrium pricing must be determined locally. The transfer problem and the diversity of markets create additional calculation requirements for an already strained energy industry.

- An error in one's own ability to predict its production can be very costly, as real-time transactions must compensate for shortages or surpluses of production at completely different prices. A varying amount of electricity must be wasted to maintain the market balance, says Andreas Wiklander.

Stress for the staff

This has forced energy companies to diversify.

- If we only sell our hydropower energy, we will compete with solar cells and wind power, which have completely different investment costs. We only make money when it's not windy or when it's cloudy, but at the same time we have to produce hydropower all year round due to all the water laws. Therefore, it is important that we broaden and complement with other products. But then optimization problems arise. The players who continue to only sell planned hydropower will find it very difficult with profitability in the future.

Production planning becomes more labor-intensive and the companies become extremely dependent on their own competence, which is already stressed to the extreme today. This increases both costs and risks.

- The responsibility becomes heavy and burdensome for the staff. The human brain is not always enough.

AI is a necessity

For energy companies, AI is a matter of survival.

- The actual optimization towards different markets must be automated and in the future based on some kind of machine learning. With the help of good AI models, you can reduce the pressure on your own staff and create new opportunities for optimized production and more efficient markets, says Andreas Wiklander and concludes:

- But it is important to optimize and use the technology in the market where it is most useful. Then the result will be good for the energy companies, consumers and for society. We can optimize more efficiently between different markets and at the same time manage our water resources better. The absolute biggest benefit for us is to always be able to sell energy optimally to the right market at the right time.

The Autonomous Production