Decarbonization of energy-intensive industry through smart sector coupling with AI-based probabilistic forecasting and operations management (DeepCarbPlanner).

Focus topic:	Resource efficiency
Type of funding:	Project funding programmes
Programme:	CZS Transfer
Funded institution:	Hochschule Konstanz – Technik, Wirtschaft und Gestaltung

With the help of machine learning methods, measures for effective and economical emission reduction are being researched in the project. The digital twin of a production process is used to show ways to achieve climate-neutral production.

Goals

The industrial sector is responsible for around 23 % of greenhouse gas emissions in Germany and is therefore of crucial importance for meeting climate targets. The transformation to more climate neutrality that is thus necessary requires a step-by-step conversion of the processes and their operations in the energy-intensive industries.

The aim of the project is to use machine learning methods to find measures for effective and economical emission reduction, for example through sector coupling between renewable energy generation and storage technologies.

Using the example of a cooperation partner, the conversion to CO2-neutral processes as well as an efficient utilization of CO2-neutral energy sources will be demonstrated. For this purpose, a digital twin of the production process will be developed and used to show ways towards climate-neutral production. In the process, probabilistic predictions based on machine learning as well as algorithms for operations management will be developed.