The production of chemicals consumes huge amounts of energy. In fact, it accounts for 20 percent of Europe's total commercial energy demand. Any reduction in this energy consumption would have a beneficial effect both on the environment and on the budgets of chemical companies. To achieve this goal, a team led by Dr. Michael Bortz and Professor Karl-Heinz Küfer of the Fraunhofer Institute for Industrial Mathematics ITWM developed a model-based toolbox. The algorithms combine machine learning methods with physical contexts to make the processes they model as realistic as possible. The result is that energy savings in the double-digit percent range have already been achieved in existing production plants.
The chemical giant BASF SE as well as the Swiss chemical and pharmaceutical company LONZA Group AG see this as an immense benefit. At BASF SE, the toolbox is available to hundreds of chemical engineers. But it’s use is not limited to the chemical industry. It can be used to advantage wherever processes with a large number of influencing factors have to be controlled.