Abstract
This chapter designs intelligent data analytic approaches for predicting dissolved oxygen concentration in river utilizing extremely randomized tree versus random forest, MLPNN and MLR. Dissolved oxygen concentration (DO) in river, lake and stream can be measured directly in situ. However, mathematical models based on intelligent data analytic technique can provide a reasonably good alternative by linking several water quality variables to the concentration of DO at different time scale. Recent studies conducted worldwide have successfully demonstrated that models using intelligent data analytics contribute to accurately estimate dissolved oxygen with high precision. Here, we applied the extremely randomized tree (ERT) to develop a robust and computationally simple model for predicting dissolved oxygen concentration in river. Results obtained using the proposed ERT were compared to those obtained using the random forest (RF), the multilayer perceptron neural networks (MLPNN) and the standard multiple linear regression (MLR). The proposed models were developed using several inputs variables, e.g. water temperature, specific conductance, water pH and phycocyanin pigment concentration. Several inputs combinations were considered and compared to find the best inputs variables for predicting DO. All the proposed models were applied and compared using data collected from two rivers located in the USA. The accuracy of the models was evaluated using coefficient of correlation (R), Nash–Sutcliffe efficiency (NSE), root mean squared error (RMSE) and mean absolute error (MAE). Results were evaluated based on several input combinations and they showed that the RF provided the most effective estimation of DO concentration amongst the all the proposed models, while the ERT was ranked in the second place slightly less than the RF, the MLPNN ranked thirdly and the MLR model provided the worst accuracy.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.