Photo: ZSW

ZSW AI team develops interactive platform for companies

Artificial intelligence is being used more and more, especially in the energy transition industry. Self-learning methods help, for example, to better predict wind and solar energy feed-in or to optimize the development and production of fuel cells, batteries and e-fuels. To ensure that small and medium-sized companies are not left behind by the big players in the industry, they need future-proof technologies. The Centre for Solar Energy and Hydrogen Research Baden-W├╝rttemberg (ZSW) has therefore developed an innovation platform where companies can test for themselves how they can use artificial intelligence (AI) to make their processes from product development to business operations more future-proof and economically viable. Ideas and concepts for this were developed by the ZSW as part of the “AI Lab for Renewable Energies”, which was financially supported by the Baden-W├╝rttemberg Ministry of Economics.

“Our AI platform covers a broad range of applications: Whether new technologies for the energy transition, whether climate-neutral production processes or applications – this is what our innovation platform stands for. Small and medium-sized enterprises – and especially startups – along the entire value chain in the field of renewable energies can benefit from it,” explains Anton Kaifel, team leader of the AI and Machine Learning department at ZSW.

First test, then decide

Using the ZSW service is simple: companies register on the website, upload their data to a protected data room and can use software and infrastructure to independently train AI models with their data. “Thus, companies have a very low entry threshold and can use our AI PlayGround tool to test whether and how artificial intelligence can be integrated into their company and whether it is worthwhile for them – in other words, test-before-invest,” explains Dr. Frank Sehnke, Data Scientist for AI at ZSW. He and the ten-member AI research team at the ZSW know the requirements and needs of companies very well. They gained experience in the “AI Lab for Renewable Energies (KILEE)”, where companies were able to use AI methods to optimize production processes and develop new products and services. The project started in April 2020 and has now been successfully completed. “Thanks to the AI Lab and other research projects, the technological maturity of our platform is very high. Machine learning can be used to improve products, processes and services and to develop new business models. This is done using existing or continuously collected data,” says Frank Sehnke.

Several companies from different industries in the Southwest participated in the regional AI lab, including weather and energy forecasting service providers and companies from the wind and solar industries. The Freiburg-based high-tech start-up greenventory, which specialises in software-supported solutions for the inventory and analysis of distributed energy systems, also benefits from the tailor-made offers from the ZSW digital laboratory. “For us as a young start-up company in the energy sector, the AI Lab was and is a very great help. The experts at the ZSW have shown us concrete possible uses of AI and exemplary applications with our data. I think it is very good that an AI platform has now emerged from the lab. Everyone should be able to use the enormous potential that AI technology offers. The platform offers low-threshold access for the use of AI in one’s own company,” says greenventory co-founder Dr. Kai Mainzer.

Analysis tool for companies nationwide

While the focus of the KI-Lab was on companies from the southwest, the platform with its analysis tool is now available to companies from all over Germany. “We want to give companies a competitive edge with our digital service by helping them achieve greater effectiveness, efficiency and agility. The dynamic development of the market requires flexibility and an AI solution that supports new business models,” says Anton Kaifel.

Companies that sign up on the platform go through a process. First, areas of application for AI can be determined in the respective company. From this, an AI readiness level is determined and further steps are discussed. Then, a proof-of-concept (PoC) application with AI will test the data and develop initial AI models using the platform. Tutorials provide guidance in the handling of the AI platform. If AI models based on a PoC are to be used by the companies for process optimization for monitoring and predictive maintenance of plants or for new services and products, the ZSW supports the companies to develop individual applications and software. This enables companies to significantly shorten the time horizons for new developments and to introduce their products and services to the market earlier.

With the AI platform, the ZSW primarily wants to effectively use the interfaces between the digital and energy industries, between startups and established companies along the entire value chain in the field of renewable energies, as AI has great innovative power in this area. “A study commissioned by the Federal Ministry for Economic Affairs and Energy (BMWi) shows that companies that use artificial intelligence achieve higher profits for the same turnover and create additional jobs. Through AI, companies are more likely to produce innovations and set milestones,” concludes Anton Kaifel.

The ZSW AI platform can be accessed via https://kilab-ee.zsw-bw.de.