6533b857fe1ef96bd12b4d5d
RESEARCH PRODUCT
Application of Selected Methods of Black Box for Modelling the Settleability Process in Wastewater Treatment Plant
Andrzej GawdzikBartosz SzelągAlicja Gawdziksubject
0209 industrial biotechnology020901 industrial engineering & automationEnvironmental EngineeringWaste managementChemistryBlack boxScientific method0202 electrical engineering electronic engineering information engineeringEnvironmental Chemistry020201 artificial intelligence & image processingSewage treatment02 engineering and technologydescription
Abstract The paper described how the results of measurements of inflow wastewater temperature in the chamber, a degree of external and internal recirculation in the biological-mechanical wastewater treatment plant (WWTP) in Cedzyna near Kielce, Poland, were used to make predictions of settleability of activated sludge. Three methods, namely: multivariate adaptive regression splines (MARS), random forests (RF) and modified random forests (RF + SOM) were employed to compute activated sludge settleability. The results of analysis indicate that modified random forests demonstrate the best predictive abilities.
| year | journal | country | edition | language |
|---|---|---|---|---|
| 2017-03-01 | Ecological Chemistry and Engineering S |