Answers are grounded in this paper and cite their sources.
The present study integrates experimental and intelligent models in the modeling and optimization of the iron electrode-based electrocoagulation-flocculation (EC-FLC) process for the treatment of aquafarm effluent (AFE). The modeling was done using Response Surface Methodology (RSM), Artificial Neural Networks (ANN), and Adaptive Neuro-Fuzzy Inference Systems (ANFIS), while the optimization was performed with Particle Swarm Optimization (PSO). The physicochemical properties of AFE were determined before and post-treatment to assess the effectiveness of the process. Central composite design (CCD) embedded in RSM was employed for experimental design. ANFIS model utilized the grid partition and ANN employed a multi-layer perceptron-based feed-forward architecture. The statistical analysis revealed superior prediction accuracy in the decreasing order ANFIS (R 2 : 0.9968, AAD: 0.0009, RMSE: 0.2373), ANN (R²: 0.9896, AAD: 0.0023, RMSE: 0.4388), and RSM (R²: 0.9655, AAD: 0.0067, RMSE: 0.7764), demonstrating the superiority of ANFIS model. Models’ reliability were validated by the strong correlation between the actual/predicted values of turbidity removal. The experimentally validated optimization solutions for turbidity were RSM numerical (96.71 %), RSM-PSO (97.46 %), ANN-PSO (98.83 %), and ANFIS-PSO (99.01 %). Post-treatment analysis revealed significant reductions in turbidity, nutrient content, and organic matter. These findings demonstrated the effectiveness of the intelligent tools in the modeling/optimization of EC-FLC recovery of AFE. • Blending experiments and models enhance aquafarm effluent recovery. • RSM, ANN, ANFIS, and PSO ensure accurate predictions, with ANFIS as the best tool. • Examining AFE properties before and after treatment enriches the process of understanding. • RSM, ANN, and ANFIS combined with PSO yield effective reduction strategies. • Significant post-treatment improvements establish successful pollutant reductions.
No takes yet. Share an insight, caveat, or question.
Igwegbe et al. (2024) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: