Dataset – Artificial Neural Network (ANN) modelling of wastewater BOD and COD remediation using Hydroponic Vetiver System (HVS)

The authors have reviewed 12 studies that evaluate the wastewater remediation potential of Biochemical oxygen demand (BOD) and Chemical oxygen demand (COD) by application of the Vetiver Grass (VG) under hydroponic conditions. 105 datasets of improvement of BOD and COD, each, using the Hydroponic Vetiver system (HVS) were separately tabulated (Ghosh & Sarkar, 2021). Then, HVS predictive modelling of wastewater remediation of BOD and COD was done using Artificial Neural Network (ANN).

Four neural networks (NN) were trained and their respective results are tabulated below as per the following nomenclature: • Table 1: NN1 – BOD All Data (105 datasets) • Table 2: NN2 – BOD Truncated Data (85 datasets) • Table 3: NN3 – COD All Data (105 datasets) • Table 4: NN4 – COD Truncated Data (84 datasets) The independent (x) and dependent (Y) variable denominations used for data tabulation were as follows: • Y = Removal efficiency of pollutant parameters of BOD/COD (%) • x1 = inlet or influent pollutant concentration of BOD/COD (mg/L) • x2 = initial plant density (tillers/cu.m) • x3 = hydraulic retention time (HRT) (days) • x4 = molarity of H+ ions (no.), referred henceforth as 'molarity' (mol/L) The four tables are listed subsequently.