Exploring ANFIS application based on actual data from wastewater treatment plant for predicting effluent removal quality of selected major pollutants

Efficient simultaneous multi-pollutant removal in wastewater treatment plants (WWTPs) is of critical importance in meeting increasingly stringent discharge standards. This study investigated the effectiveness of the Adaptive Neuro-Fuzzy Inference System (ANFIS) model for predicting the removal of major pollutants in a WWTP. The Parameters were screened using Principal Component Analysis (PCA), and Orthogonal Experiments (OE) were conducted to enhance ANFIS accuracy. The actual removal values and predicted values of major pollutants such as COD, BOD, NH3, and SS were compared and analyzed. The study obtained satisfactory linear results within a 95 % confidence interval, with some R2 values exceeding 0.950. Additionally, there was revealed a lack of oneto-one correspondence between predicted and actual pollutant values, for the same input value yielding different output values. This study highlighted the potential of ANFIS for pollutant removal prediction in WWTPs, yet further investigations were required to refine the model's logic and enhance accuracy for practical applications.