Artificial intelligence algorithm application in wastewater treatment plants: Case study for COD load prediction

Abstract Most modern urban wastewater treatment plants adopt a treatment technology based on an activated sludge process. Because the composition and content of pollutants in sewage cannot be known, excessive aeration and chemicals are often used to ensure that the quality of the treated sewage meets the requisite standard. Such extensive operations also result in energy wastage and potential secondary pollution. Therefore, this chapter proposes a chemical oxygen demand load prediction model based on the gradient boosting decision tree (GBDT) algorithm. Based on the prediction results, the aeration and chemical dosage can be accurately controlled to reduce energy consumption and eliminate the potential secondary chemical contamination risk.