The FAW Ulm (Research Institute for Application-oriented Knowledge Processing) was founded in 1987 as the first independent institute for artificial intelligence. Companies such as DaimlerChrysler AG, Jenoptik AG, Hewlett-Packard GmbH, Robert Bosch GmbH and several others were involved.

In the meantime, AI has found its way into many areas, be it medicine, law, marketing or computer games. The best known are machine translations, e.g. with Google Translate or Deepl, in analyzing and forecasting share price developments or handling the flood of information in search engines.

Artificial intelligence is a branch of computer science that deals with the automation of behavioral patterns, from which decision-making aids can be derived and, in the best case, autonomous processes can be continued. It is usually used when an oversized or disorganized but unmanageable amount of data needs to be managed and coordinated.

It is not always crowned with success. Amazon, for example, had to switch off its AI for evaluating applicants because the automatic evaluation system put women at a disadvantage.

And even in machine translations, there are still quite a few rough blocks that make you frown or smile when you take a closer look.

So it’s not that easy with artificial intelligence. The problem is not actually the amount of data, but rather the correct classification. Because Amazon had previously mainly hired men, the AI concluded that there was a performance deficit among women. In fact, however, less consideration was given to the fact that the low proportion of women in male-dominated professions has sociological causes.

The fundamental problem with artificial intelligence: The programming of the algorithms and the output data are only as good as the subjective work of the developers themselves who develop it and make it available. Deficits in objectivity due to individual emotions and intentions, as well as errors of interpretation and perception on the part of the developers, are taken on board by the AI, which learns from them and expands on them. Add to this a lack of knowledge about the interrelationships between things and processes (key qualifications) and the circle is complete.

AI therefore needs a lot of development time and the courage to face setbacks before it can develop into an efficient system.

Headlines such as “Artificial intelligence (AI) as a driver of the energy transition” or “How logistics is benefiting from artificial intelligence” are media hits that do not even begin to reflect the development and effort that needs to be made and the fact that costs are the first consideration before financial profitability becomes apparent.

To date, artificial intelligence has mainly been used in the energy industry for monitoring and forecasting tasks.

Smart Grid – Intelligent power grid

However, as the proportion of electricity from renewable energies increases, it is becoming clear that AI will also control energy system processes on a large scale in the future.

Artificial Intelligence (AI) – Smart Grid – Intelligent power grid – @shutterstock | monicaodo

While electricity grids with centralized power generation have dominated up to now, the trend is moving towards decentralized generation plants. This applies to generation from renewable sources such as photovoltaic systems, solar thermal power plants, wind power plants and biogas plants. This leads to a much more complex structure, primarily in the areas of load control, voltage maintenance in the distribution grid and maintaining grid stability. In contrast to medium-sized to large power plants, smaller, decentralized generation plants also feed directly into the lower voltage levels such as the low-voltage grid or the medium-voltage grid.

Development of a smart grid

An intelligent electricity grid integrates all players into an overall system through the interaction of generation, storage, grid management and consumption. Power plants (including storage) are already controlled today in such a way that the same amount of electrical energy is always produced as is consumed. Intelligent electricity grids include consumers as well as small decentralized energy suppliers and storage systems in this control system, so that consumption is balanced in terms of time and space (smart power/intelligent electricity consumption) and non-dispatchable generation systems (e.g. wind energy and PV systems) and consumers (e.g. lighting) can be better integrated.

Due to the greater proportion of renewable energies, it is becoming more important to match fluctuations in energy generation to fluctuations in energy consumption. In addition to the possibility of storing electrical energy using energy storage systems or storage power plants, electricity generation in line with demand, e.g. through hydroelectric power plants or bioenergy, the expansion of electricity grids for rapid distribution over a large area, there is also the possibility of adapting electricity consumption to the supply of electricity.

“Generating electricity from solar and wind power plants makes the supply system much more fragmented and weather-dependent than the operation of conventional power plants. In addition, consumption must be more closely aligned with the supply of electricity. The flexibility required for this cannot yet be managed with the existing infrastructure. A decentralized system can only function via digital processes in real time and automated decisions,” explains Prof. Dr. Clemens Hoffmann, Head of Fraunhofer IEE. Hoffmann sees digitalization as the basis for the next steps in the energy transition: “The coordination and decision-making processes of a decentralized renewable energy supply are extremely complex. Only artificial intelligence will make it possible to connect different systems such as electricity and heat supply as well as mobility via automated decisions on a large scale. By establishing an ecosystem for cognitive energy systems, we are advancing the application of AI in the energy sector.”

A decentralized energy system needs AI

There is already a concrete need for AI in various areas of the energy industry. For example, automated energy trading involves systems that independently identify trading strategies and trigger purchases or sales. Photovoltaic and wind power plants as well as charging stations and electrolysers can optimize their operation with AI, thereby avoiding maintenance and increasing their service life. In the grid sector, the technology is used to evaluate a wide range of information, identify critical situations and help resolve them.

The Fraunhofer IEE has been working for 15 years on artificial intelligence for forecasting weather-dependent electricity generation from solar, wind and bioenergy. An automatic trading system for the EPEX Spot electricity exchange is also being developed in Kassel.

Research for AI in the energy industry

“Artificial intelligence is a key technology for the further development of the energy transition: Moving away from a centrally organized power plant industry based on fossil fuels to an energy system based on renewable sources is a very complex process that can only be mastered through intelligent control,” said Hesse’s Science Minister Angela Dorn. “The Competence Center for Cognitive Energy Systems gives scientists space for new ideas and research approaches to innovations in the energy industry. I am delighted that we are supporting its establishment. Now it is important to combine the expertise of researchers with strong partners from industry.”

A new competence center for cognitive energy systems is therefore being established in Kassel. The research project on artificial intelligence in the energy system is looking for partners from science and industry and believes that Germany is well placed to become a global innovation leader in this field as a location for business and research. This is why the state of Hesse is funding the establishment of the new competence center, which is supported by the Fraunhofer Institute for Energy Economics and Energy System Technology IEE.

The new Cognitive Energy Systems Competence Center in Kassel is researching these fields of application for AI and is being funded by the Hessian state government with a total of 5.8 million euros between 2020 and 2022.

The K-ES

The Competence Center Cognitive Energy System (K-ES) has been established by Fraunhofer IEE since mid-2020 to research the topics of cognitive energy management, cognitive energy networks and cognitive energy system technology. The development process is scheduled to last ten years. The K-ES is set to become a national and international center for artificial intelligence in research and teaching.

The Competence Center Cognitive Energy Systems (K-ES) looks at the tasks in the energy system from an AI perspective and develops them further in the three areas of cognitive energy management, cognitive energy grids and cognitive energy system technology. “A cognitive energy system determines its own state based on available information and learns to achieve predefined goals. Artificial intelligence is not opposed to human intelligence, but is in constant communication with it and supports it. As the technology develops, both sides will change,” explains IEE project manager André Baier.

The energy industry can also build on findings from other sectors. AI is already having a lasting impact on the automotive industry, retail, insurance and the financial sector. For the energy transition with renewable energies and sector coupling, the most important areas of digitalization are smart generators and consumers, virtual power plants, smart grid technologies and the real-time energy industry.

Concepts and applications for the economy

The concept for setting up the K-ES was developed by the Fraunhofer IEE. The initiative goes back to an agreement from the coalition agreement of the Hessian state government. The development phase has now begun. The primary aim is to create an ecosystem for innovation and form a community of experts. The new competence center will be part of the Fraunhofer IEE campus in Kassel, which is currently under construction, and will complement the research spectrum for the transformation of energy systems.

The first step is to set up premises and the IT infrastructure with a cloud system. This will be followed by the creation of a digital platform through which partners from business and research can exchange information. The focus of the start-up phase will be on recruiting scientists and building up expertise. “Our aim is to connect scientists who have a common goal in mind, regardless of where in the world the experts are based,” says Baier.

Until the planned official establishment of the competence center, the focus will also be on acquiring partners and application projects from industry. After all, close links with the energy sector are part of the concept: K-ES services for energy companies include consulting and concept studies, prototypes and turnkey systems. “We welcome applications from researchers and companies alike, because an ecosystem like this thrives on networking between theory and practice,” emphasizes Hoffmann.

The goal: A community of international renown in Germany

Over the next ten years, it is planned that around 100 experts at K-ES will be working in the disciplines of data science, advances in machine learning, recommender systems and digital innovation management. There are currently 15 employees at Fraunhofer IEE working in these fields. The aim of the new institution is to become one of the leading communities for AI in the energy industry in Germany.

In order to take account of the highly international nature of AI research, the competence center also offers guest scientists from all over the world the opportunity to participate. “Thanks to the special training infrastructure, appropriate hardware and software as well as a comprehensive model and data pool, we can conduct AI research for the energy system efficiently and across locations,” says Christoph Scholz, Scientific Director of K-ES, explaining the available opportunities.

Intensive work is being carried out on the development of AI worldwide. Germany has so far spent significantly less on corresponding research than its competitors, the USA and China. As part of the German government’s Corona Future Package, €5 billion is now to be invested in AI by 2025. “As a business and research location, Germany is well placed to become a global innovation leader when it comes to AI in the energy system. To achieve this, it is important that all stakeholders work together to advance the topic,” says Hoffmann.

Cognitive systems

A cognitive system is a digital system with interfaces between the digital world and the environment that can perceive and understand things and draw conclusions and learn from them. Cognitive systems are able to independently develop solutions for human tasks. They can interact and cooperate with other digital systems, interpret contexts and are adaptable.

Cognitive systems are being used in an increasing number of areas and represent, for example, the basic technology for self-driving vehicles, intelligent personal assistants, Industry 4.0 and the Internet of Things. A typical feature of such systems is that they can process large amounts of data in a short space of time and are embedded in a higher-level system (system of systems). Tens of billions of euros have been invested in this technology worldwide up to 2020.

A cognitive system can independently determine its own state and that of its assets based on available information and learn to achieve predefined goals autonomously thanks to its ability to adapt. Cognitive energy systems are a key technology for the energy transition. Applications in the electricity industry can be found in the area of grid management and the management of generation and consumption.

Within the ecosystem for cognitive energy systems, access to AI will be made easier for the various market roles. The tasks of system operators, metering point operators, balancing group managers and direct marketers are automated to such an extent that they run independently. The “energy avatar” model (see above) illustrates how easy it is for a “home builder” to participate in the energy market with their solar installation if all processes are automated. The energy avatar is currently being developed in collaboration between the Fraunhofer Institutes IEE and IOSB-AST.

A close connection with the energy sector is part of the concept: K-ES services for energy companies range from consulting and concept studies to prototypes and turnkey systems. The ecosystem thrives on networking between theory and practice.