Computational-Based Approaches for Predicting Biochemical Oxygen Demand (BOD) Removal in Adsorption Process

Predicting the adsorption performance to remove organic pollutants from wastewater is an essential environmental-related topic, requiring knowledge of various statistical tools and artificial intelligence techniques. Hence, this study first develop a quadratic regression model neural network (ANN) for predicting biochemical oxygen demand (BOD) removal under different conditions. Nanozero-valent iron encapsulated into cellulose acetate (CA/nZVI) was synthesized, characterized by XRD, SEM, EDS, used as efficient adsorbent BOD reduction. Results indicated that medium pH time should be adjusted around 7 30 min, respectively, maintain highest efficiency 96.4% at initial concentration C o = 100 mg/L, mixing id="M2"> rate 200 rpm, dosage 3 g/L. An optimized ANN structure 5–10–1, with “trainlm” back-propagation learning algorithm, achieved predictive ( id="M3"> R 2 : 0.972, Adj- id="M4"> 0.971, RMSE: 1.449, SSE: 56.680). Based on sensitivity analysis, relative importance factors could arranged id="M5"> pH > adsorbent dosage time ≈ stirring speed . A developed visualize impacts efficiency, optimizing 7.3 46.2 min. The accuracy models in approximately comparable. these computational-based methods further maximize CA/nZVI material removing applicability modeling techniques would guide stakeholders industrial sector overcome nonlinearity complexity issues related process.