Algorithm-based trading technology, already established in financial markets, is now making its way into the world of energy trading. But uniting the requirements of trading automation with the special demands of the energy industry is no easy task. In a four-part series Wolfgang Eichberger, Co-Founder of VisoTech, shares the challenges of developing a fully automated solution for intraday power and gas trading. If you’re considering developing your own algo-trading solution, you’ll want to read this! In this first part, he looks at the evolution of energy trading technology and the drivers of change, and explains the differences between financial and energy trading.
Back in 2002 when VisoTech first began development of Periotheus, our software suite for energy process automation, short-term energy trading was still a highly manual process. Planning was primarily managed across multiple spreadsheets. Processes were often carried out on paper. Communication occurred over the phone or even by fax. The business processes in the energy industry were still in their infancy. A playground for software providers – just a little effort could deliver massive value.
Now, more than 15 years later, we’re living in another world – the requirements have increased dramatically. Liberalisation and the rapid growth of renewable energy led to an explosion in the volume (Fig. 1) and volatility (Fig. 2) of short-term energy trading. Traders faced completely new challenges. With falling energy prices, and a resulting shrinking of margins, new, cost-effective solutions are needed. The magic word is digitalisation! However, the attempt to apply systems from the finance world to short-term energy trading promptly failed. The energy market plays by its own rules.
Why solutions from financial trading won’t work
Financial trading relates to distinct, abstract products. Energy trading, on the other hand, revolves around a physical product that must be delivered. Trading decisions are highly dependent on external factors such as weather, power plant breakdowns, transmission problems… so the market is only a reflection of the physical reality. This makes interfaces much more important. Data comes into the system with different structures that have to be processed, and in the end results have to come back to do something physical – to manage power plants and supply customers.
Unlike the financial world, energy traders need to understand where the energy comes from, and the same goes for an automated system. The entire process chain must be represented. Decisions must be made: what amounts from which quarter hour must be delivered from which facility. And also unlike the financial markets, which have regular opening and closing times, this has to occur around the clock, to ensure supply and demand are balanced and to keep the grid operating smoothly.
Up to now we’ve invested over ten person years, including three complete rewrites.
For these reasons, in 2015 we began work on an automated trading solution for intraday energy markets. At first this seemed easy, but it quickly snowballed into a massive undertaking. Many of our customers have come to the same conclusion; after trying to develop their own system, they switched over to our solution. Up to now we’ve invested over ten person years, including three complete rewrites and several major changes. This is in addition to development of the related upstream and downstream functions of the Periotheus Suite, such as scheduling and power plant management.
As they say, it takes a village to raise a child. The same goes for a complex product. We couldn’t have solved the great challenges of interfaces, trading algorithms and performance without the close cooperation of customers and exchange partners.
Stay tuned next week for part two, in which Wolfgang shares one of the most challenging aspects of developing an automated energy trading solution.
Register for our newsletter and receive monthly updates about our company, products and events.