How can machine learning improve waste-to-energy plant operation

Alexandra Pehlken, Henriette Garmatter, Lisa Dawel, Fabian Cyris, Hendrik Beck, Fenja Schwark, Roland Scharf, Astrid Nieße
This paper deals with the integration and the role ofwaste-to-energy plants. The hygienic and safe disposal of wasteis a central aspect of human infrastructure. It is a prerequisitefor preventing the spread of disease in society and is a veryimportant issue today as it has been in the past. While serving aswaste disposal in the past, today, waste incineration has changedto waste-to-energy plants, in addition to keeping the first goalof waste reducing and sanitation. Therefore, the highest energyharvesting is not the primary goal of the plant. The plant oper-ation is difficult to control related to the stochastical variationof the properties of municipal waste due to its heterogeneousnature. Hence, it is expected that the optimization of waste-to-energy plants will benefit significantly if any applied methodmay handle the stochastic properties well. The work presentedhere aims to provide insights into a novel approach to developa new method for enhancing the performance of the waste-to-energy plant related to blast cleanings to prevent build-upof particles. This also ensures that the overall performance ofthe plant improves. Artificial Intelligence was applied to sensoremission data directly from an operating real incinerator andwith machine learning the data shows the effects of the blastcleanings related to typical plant properties. The results of thispaper give first indications that forecasting of the incinerationprocess is possible. With these results, the plant operator couldhandle the plant efficiently with the least performance reduction.Index Terms—waste-to-energy, machine learning, artificial in-telligence, regression, XGBoost
Abfallverbrennungskessel 4.0 - Digitalisierung und ganzheitliche Betriebscharakterisierung von Dampferzeugern in Abfallverbrennungskraftwerken