Dissolved Oxygen Prediction of the Ciliwung River using Artificial Neural Networks, Support Vector Machine, and Streeter-Phelps

Abstract

Evaluation of Ciliwung river water quality can be done by analyzing the distribution of dissolved oxygen (DO). The purpose of this research is to analyze the environmental parameters that affect the distribution of DO, by carrying out predictive modeling to estimate the distribution of DO in the Ciliwung River. The research data used primary data and secondary data, some of which were obtained from previous studies. The water quality parameters used are DO, temperature, biochemical oxygen demand, chemical oxygen demand, power of hydrogen, and turbidity. The dataset used has a missing value of 28.8%. To optimize the model results, preprocessing is carried out using a machine learning approach, namely comparing support vector machine (SVM), artificial neural networks (ANN), and linear regression. The three models were compared to predict DO, the results of performance evaluation of the SVM, ANN and Streeter-Phelps models had RMSE values of 0.110, 0.771, and 0.114.