AI-based fire power control makes waste incineration more efficient

The combustion of waste to generate heat, steam, and electricity is a highly complex process as the fuel is nonhomogeneous. This may lead to faults or suboptimal combustion.

By means of digital methods, the quality of the feed as well as of the production process and thus the efficiency of the entire plant can be increased significantly.

The software solution PiT Navigator ensures a continuous feed and combustion process
The solution consists of a combination of video and thermograpic camera in a refractory housing and is oriented from the rear towards the endmost combustion zones. The recorded thermographic images are digitally evaluated. They allow to determine the current temperatures in the furnace, which in turn have an impact on the speed of the feed system, the fan control, and ultimately the emission values. The process runs automatedly owing to the use of AI components.

All process values are improved
The goal of this furnace capacity control is to keep the steam power that drives the turbine to supply steam to the district heating network or to adjacent industries as constant as possible. Thus if the gross calorific value is insufficient, the speed of the feed will be increased. If the combustion process does not run optimally, which manifests itself e.g. in the emergence of hotspots or a fuel imbalance between the right and the left part of the grate system, the speed will be adjusted as well and the fan will be regulated accordingly. This, in turn, helps to prevent emission peaks of carbon oxides and nitrous oxides and to reduce their emission in general.

Technologically unique solution to date
The combination of a steam prognosis two to five minutes in advance by means of neural networks, enhanced process control, and intelligent data processing is technologically unique to date. It still allows to manually influence the process at any time. Both, the intelligent control of the combustion process and the experience of the staff members in the case of manual influence, lead to an optimal control of the mutually influencing target variables in the waste incineration process and the steam generation.

Success variables achieved in a project in Spain:

• -25 % reduction of steam fluctuations (achieved: 45 %)
• +3% increase in steam quantity
• +3% higher waste throughput (by stabilizing the steam power)