Scholarlink

Wastewater treatment is a critical solution that decreases environmental effects by removing organic waste and producing a sustainable water source. However, the fundamental issue in running a wastewater treatment plant is the excess chemical oxygen demand concentration due to sludge thickening. The quality of the WWTP's effluent water will decrease due to sludge thickening, which will also bring about inadequate sludge loss and settling. Consequently, one of the crucial problems that must be resolved to guarantee process safety and enhance effluent quality is the accurate identification of sludge bulking. Machine learning, a branch of artificial intelligence (AI), can upend societies, economies, and environments globally by enabling systems to get knowledge straight from data, examples, and experience rather than from pre-established rules. This research utilised sophisticated machine learning such as CatBoost, Random Forest and Decision Trees and improved efficiency of wastewater treatment facilities. Also, the SHapley Additive exPlanations (SHAP) explainer was employed to interpret machine learning models, offering insights into complex model predictions by elucidating the impact of each feature on the model's output. The mode fit achieved was more than 90% in predicting Chemical Oxygen Demand (COD), thereby lowering operating costs, improving effluent quality, and effectively controlling sludge bulking.

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

Explainable Models for Prediction of Chemical Oxygen Demand in Industrial Wastewater Treatment Plant Facility in Delfzijl, The Netherlands

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Ismaila Abubakar Bobboi,
  • Zubaida Said Ameen,
  • Dilber Uzun Ozsahin,
  • Auwalu Saleh Mubarak

摘要

Wastewater treatment is a critical solution that decreases environmental effects by removing organic waste and producing a sustainable water source. However, the fundamental issue in running a wastewater treatment plant is the excess chemical oxygen demand concentration due to sludge thickening. The quality of the WWTP's effluent water will decrease due to sludge thickening, which will also bring about inadequate sludge loss and settling. Consequently, one of the crucial problems that must be resolved to guarantee process safety and enhance effluent quality is the accurate identification of sludge bulking. Machine learning, a branch of artificial intelligence (AI), can upend societies, economies, and environments globally by enabling systems to get knowledge straight from data, examples, and experience rather than from pre-established rules. This research utilised sophisticated machine learning such as CatBoost, Random Forest and Decision Trees and improved efficiency of wastewater treatment facilities. Also, the SHapley Additive exPlanations (SHAP) explainer was employed to interpret machine learning models, offering insights into complex model predictions by elucidating the impact of each feature on the model's output. The mode fit achieved was more than 90% in predicting Chemical Oxygen Demand (COD), thereby lowering operating costs, improving effluent quality, and effectively controlling sludge bulking.