Optimized control of waste incineration plants

Incinerating waste in waste incineration plants can feed local and district heating networks and generate electrical energy. It is important that the district and local heating networks are supplied with steam that is as constant as possible and that the electrical energy is also fed into the grid almost constantly. Furthermore, the waste should also be incinerated as emission-free and completely as possible.

The difficulty here is that the waste produced varies greatly in its composition, which leads to widely varying heating and emission values. For this reason, it is also difficult to fully automate firing rate control, meaning that conventional control concepts are sometimes unable to cope with this task as well as a well-trained plant expert (operator).

As part of the research project “Optimized control of waste incineration plants”, a visual process monitoring system for waste incineration plants is being expanded to include AI-based predictions of the heating and emission values of a waste batch and a classic firing rate control system is being optimized using this additional information. Project partner SWB Entsorgung GmbH is making the Mittelkalorikkraftwerk (MKK) in the port of Bremen available for this purpose. This waste incineration plant offers a good starting point, as the second project partner, Bremen-based start-up WasteAnt, has already implemented an image processing system with AI-supported image analysis there.

The first goal of this project is therefore to design an observer for determining the calorific value of the current waste batch and the emissions produced. WasteAnt will then use this information to train its AI so that it can predict the heating and emission values of a waste batch based on images. In the final step, concepts will be developed to optimize the existing firing rate control in the MKK using the predicted heating and emission values.