Gross parameters prediction of a granular-attached biomass reactor by means of multi-objective genetic-designed artificial neural networks: touristic pressure management case

[object Object]The Artificial Neural Networks by Multi-objective
Genetic Algorithms (ANN-MOGA) model has been applied
to gross parameters data of a Sequencing Batch Biofilter
Granular Reactor (SBBGR) with the aim of providing an effective
tool for predicting the fluctuations coming from touristic
pressure. Six independent multivariate models, which
were able to predict the dynamics of raw chemical oxygen
demand (COD), soluble chemical oxygen demand (CODsol),
total suspended solid (TSS), total nitrogen (TN), ammoniacal
nitrogen (N-NH4
+) and total phosphorus (Ptot), were developed.
The ANN-MOGA software application has shown to be
suitable for addressing the SBBGR reactor modelling. The R2
found are very good, with values equal to 0.94, 0.92, 0.88,
0.88, 0.98 and 0.91 for COD, CODsol, N-NH4
+, TN, Ptot and
TSS, respectively. A comparison was made between SBBGR
and traditional activated sludge treatment plant modelling.
The results showed the better performance of the ANNMOGA
application with respect to a wide selection of scientific
literature cases.