Prediction Of BOD And COD For Sewage Treatment Plant Using Artificial Neural Network Approach

Abstract

In this study, the knowledge base of a genuine wastewater treatment plant was acquired using a Neuro Vector Machine modelling approach, which was subsequently applied as a process model. The study shows that ANNs integrated with Support Vector Machine (SVM) are capable of accurately capturing the characteristics of plant functioning. The trained ANN plant model is included into a computer model. Utilizing plant scale data collected from a nearby wastewater treatment plant, the designed program is put into use and evaluated. For plant operators and decision-makers, it serves as a useful performance assessment tool. When employing COD as an input in the crude supply stream, the proposed model accurately predicted the effluent stream's biological oxygen demand (BOD), chemical oxygen demand (COD), and total suspended solids (TSS). One may argue that combining three crude supply inputs—BOD, COD, and TSS—rather than just one produced better model predictions than using just one crude supply input. The proposed approach for the Vitthalwadi STP data is done and presented via a graphical user interface and attained an Accuracy of about 95.6%.