Novel PCA-driven extreme machine learning for comprehensive modelling of metropolitan wastewater treatment systems
Vini Antony Grace, Ghadah Aldehim, Nuha Alruwais, Prabakar T.N
Abstract
This study models metropolitan wastewater treatment plants (MWWTPs) in Kolkata using an Extreme Learning Machine (ELM) combined with Principal Component Analysis (PCA). PCA reduces input data dimensionality, while ELM enhances predictive accuracy. The proposed PCA-ELM model significantly outperforms standard ELM and traditional models like Multiple Linear Regression (MLR) and Multi-Layer Perceptron (MLP), improving accuracy for key parameters such as BODeff (60.1 %), CODeff (92.3 %), TNeff (86.5 %), and TPeff (72.5 %). These results demonstrate ELM's effectiveness for wastewater treatment modeling and sustainable management.
Topics & Concepts
Metropolitan areaWastewaterSewage treatmentExtreme learning machineComputer scienceEnvironmental scienceArtificial intelligenceMachine learningProcess engineeringEngineeringEnvironmental engineeringGeographyArtificial neural networkArchaeologyMachine Learning and ELMWater Quality Monitoring TechnologiesWater Systems and Optimization