Forecasting Effluent Biochemical Oxygen Demand in Sewage Treatment Plants Using Machine Learning and User-Friendly Interface

Efficiency of a system in a sewage treatment plant (STP) is significant in providing high quality of treated water to be discharged for the usage of surrounding neighborhood. However, the problems in measuring and monitoring the water quality in the treated wastewater or effluent water in real time has made it difficult to maintain the efficiency and preserve the energy of the STP. Therefore, this study purposes a graphical user interface (GUI) that has been embedded with a machine learning model to predict effluent parameters in real time. In this study, artificial neural network (ANN) and support vector machine were developed to predict biochemical oxygen demand (BODeff) using several effluent variables. Both models were evaluated using correlation coefficient (R), root mean square error (RMSE), determination of coefficient (R-2), and mean square error (MSE). Based on the results, ANN model has outperformed SVM model by achieving the value of MSE and RMSE < 0.02 while R and R-2 near to the value of 1.00. Therefore, the ANN model has been inserted into the GUI as ANN model is the most optimum model to predict BODeff in real time. The GUI-based app also performed well and able to predict the parameter with great accuracy.