Gross parameters prediction of a granular attached biomass reactor through evolutionary polynomial regression

• Prediction of granular biomass reactor response was studied.
• Evolutionary Polynomial Regression (EPR) was applied to a granular biomass reactor dataset.
• EPR provided highly reliable models for estimating granular biomass reactor behaviour.
• Using models can provide valuable information to address the tourist stress issue.

Heavy fluctuations in wastewater composition, such as those typical of tourist areas, can lead to a deterioration in treatment plant performance if no action is taken in advance. Mathematical modelling, applied to treatment plant performance prediction, can provide valuable information to address the stress issue. The present study shows that the evolutionary polynomial regression methodology (EPR) is able to predict the performances of an attached granular biomass system so that it is possible to make the necessary operating changes in advance, avoiding deterioration in the quality of the effluent discharged. The present paper shows the results of EPR application to gross parameters of a granular attached biomass reactor. For each parameter, a model capable of predicting the effluent value was assessed, based on the knowledge of the influent characteristics. Coefficients of determination values (CoD) obtained during the models validation phase, can be said to be more than satisfactory, varying between 84.2% and 94.6%. Moreover, the applied tests showed typical behaviours commonly found when observed and predicted values are quite similar. This paper reports the first application attempt for modelling this kind of emerging treatment system and gross parameters.