What does artificial intelligence mean for energy traders?
AI is one of the top trends in the energy industry. But is it just another buzzword that will fade away after a few over-hyped pilot projects? The answer is clear: no! AI is clearly here to stay, and it has the potential to revolutionise industrial processes.
The enabling technologies needed for AI are rapidly evolving. Industry giants such as Microsoft, Google and IBM are investing billions in the further development of AI systems. Often the results from such research are made available as open source projects, further accelerating the spread of such systems. An example of this is TensorFlow, a system for AI applications originally developed by Google that can now be freely used by anyone.
AI is clearly here to stay, and it has the potential to revolutionise industrial processes.
So what can all these technologies accomplish? To put it briefly: prediction and optimisation. Highly granular weather forecasting – such as IBM’s Deep Thunder solution with ranges under a kilometre in 15-minute periods – enable much more accurate prediction of solar and wind energy supply as well as energy demand.
This in turn makes it possible to deliver just the right amount of power when and where it is needed. The operation of power plants of all kinds – physical or virtual, traditional or renewable – can be optimised, along with the grid itself. With smart meters and AI-powered demand side management, demand can also be shifted to the times of highest supply. In the end, all of this translates to cost savings and increased profit, combined with greater energy efficiency and lower emissions. A win-win solution for everyone, right?
Is human intelligence still needed?
Thank you, you’ve been replaced by artificial intelligence! Currently many trading professionals in the energy industry are afraid of hearing this. Rightly so? One thing is clear: the energy world is going through a period of drastic change. And it’s happening at an uncomfortably fast pace for this traditionally conservative industry. No surprise, because the rapid pace of digitalisation is forcing the industry to accelerate. An industry on speed. This has become incredibly obvious in the energy spot markets. Year after year, EPEX SPOT, Nord Pool and even the natural gas exchange PEGAS are seeing record growth in short-term trading. Meanwhile, energy can be traded just minutes before delivery.
Yet another issue that short-term trading can alleviate is the continuing drop in prices. Shrinking margins have already rendered countless business plans obsolete. Even newly built fossil-fuel power plants have been shut down, in order to avoid burning money. And to add more fuel to the fire, prices are becoming more volatile. Long-term predictability is a thing of the past.
Algo-Trading Systems Booming
Fortunately, more and more processes can be automated. In short-term markets, algo-trading systems are booming and are well-equipped to deliver the necessary speed. Already standard in the financial industry, they are now in high demand in the energy industry.
The new currency of the energy industry is know-how.
New opportunities are emerging, based on intelligent systems that automate complete process chains from data management to trading and scheduling. By learning how the market works and changes over time, AI-based trading algorithms can discover complex patterns a human might never see and close smarter, more profitable trades in the blink of an eye.
So now back to the original question – is human intelligence still necessary? Naturally! Artificial intelligence systems are – as the name implies – only artificially intelligent. These systems first have to be trained, like a little dog. Then they are only as intelligent as the trainer can make them. As a result, the knowledge of energy traders is more in demand than ever. The new currency of the energy industry is know-how. Or why is there still no Uber or AirBnB for energy trading? Easy: because it’s much more complicated than taxi dispatching or hotel booking. So energy traders are well advised to use their own intelligence in the trading race and stay a step ahead of the competition.
Wolfgang Eichberger, Co-Founder, VisoTech
Opinion piece published in: EUROFORUM Conference Newsletter “Intraday Trading – Power” September 28, 2018; © 2018 EUROFORUM Deutschland GmbH, Düsseldorf.
You can also see the original article (in German) on the EUROFORUM website.
About the EUROFORUM Conference “Intraday Trading – Power”
As an event organiser, EUROFORUM engages proven experts from business, science and politics and creates forums for knowledge transfer and exchange of views. High-quality conferences, seminars and conventions highlight the latest trends and provide up-to-date information for practical business management. In addition, executives make valuable contacts with industry decision-makers.
The conference “Intraday Trading – Power” on December 6-7, 2018, in Düsseldorf is focused on important topics in the energy business, such as
- Commercialisation opportunities on the intraday market
- Market access and prerequisites for trading
- Experiences with XBID – is cross-border trade working?
- Blockchain, peer-to-peer, local markets
- Artificial intelligence in energy trading
- Automation and algorithms: speed makes the difference!
- Experiences with algo-trading
As the supplier of Europe’s leading software solution for fully automated power and gas spot trading, we have been invited to share our knowledge. On December 7 at 11:35 VisoTech Co-Founder Johannes Püller will give a presentation on the subject of “Artificial intelligence for intraday trading: power plant optimisation on speed”.
Register for our newsletter and receive monthly updates about our company, products and events.