Scholarlink

The pollution of river water despite the presence of sewage treatment plants (STPs) is a great concern. The study on the selection of secondary biological treatment and its constant monitoring for a typical sewage is very necessary and therefore this study focuses on setting up four lab-scale biological reactors—suspended growth and attached growth, both with seeding and non-seeding and identified the best method to treat sewage from Imphal, Manipur. The parameters considered for the study include Biological oxygen demand (BOD), Chemical oxygen demand (COD), pH and total suspended solids (TSS) evolved out of sludge biomass. The correlation between the parameters was evaluated, and different machine learning (ML) algorithms such as Decision Tree (DT), Extreme Gradient Boost (XGB), Extra Tree (ET), K-Nearest Neighbor (KNN), Random Forest (RF), and Support Vector Machine (SVM) were then used to predict BOD, COD, and TSS. Comparatively, Random Forest and XGBoost performed well, with R2 values exceeding 0.95, emphasizing the promise of ML in optimizing wastewater management processes. This study highlights the potential of machine learning algorithms with Extra Trees emerging as the most effective model, achieving an R2 of 0.99 across all parameters.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine Learning-Based Performance Evaluation and Optimisation of Biological Reactors for Sewage Water Treatment

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Sudhakar Ningthoujam,
  • Supriya Thoudam,
  • Potsangbam Albino Kumar

摘要

The pollution of river water despite the presence of sewage treatment plants (STPs) is a great concern. The study on the selection of secondary biological treatment and its constant monitoring for a typical sewage is very necessary and therefore this study focuses on setting up four lab-scale biological reactors—suspended growth and attached growth, both with seeding and non-seeding and identified the best method to treat sewage from Imphal, Manipur. The parameters considered for the study include Biological oxygen demand (BOD), Chemical oxygen demand (COD), pH and total suspended solids (TSS) evolved out of sludge biomass. The correlation between the parameters was evaluated, and different machine learning (ML) algorithms such as Decision Tree (DT), Extreme Gradient Boost (XGB), Extra Tree (ET), K-Nearest Neighbor (KNN), Random Forest (RF), and Support Vector Machine (SVM) were then used to predict BOD, COD, and TSS. Comparatively, Random Forest and XGBoost performed well, with R2 values exceeding 0.95, emphasizing the promise of ML in optimizing wastewater management processes. This study highlights the potential of machine learning algorithms with Extra Trees emerging as the most effective model, achieving an R2 of 0.99 across all parameters.