1. Title, Journal and Authors

Title: A comparative analysis among computational intelligence techniques

for dissolved oxygen prediction in Delaware River

Journal: Geoscience Frontiers

Authors: Ehsan Olyaie, Hamid Zare Abyaneh, Ali Danandeh Mehr

 

2. Summary

In this study, various computational integration techniques (e.g., MLP, RBF, LGP, SVM) were compared to estimate DO concentrations. To achieve this objective, Delaware River Place in the United States was used to develop various models investigated in this study.

In this paper, the performance of SVM 0.99, LGP 0.96, MLP 0.91, and RBF 0.81 were shown through the coefficient of determination, and as a result, the SVM model had higher prediction accuracy compared to other models.

 

3. Originality and Creativity

 

 

4. Application to research

In this paper, a total of five combinations (T, T+pH, T+Q, T+pH+EC, T+pH+Q, T+pH+Q, T+pH+Q+EC) were created by combining pH, electric conductivity (EC), temperature (T), and river discharge volume (Q) as INPUT values.

R values, RMSE values, MARE values, and NS values were compared.

The SVM had the lowest RMSE, MARE values, and the highest R, NS values.

 

5. Contact

Ma, Kyoung Rim / M.S. program

 

Environmental Systems Engineering Lab.

School of Earth Sciences and Environmental Engineering

Gwangju Institute of Science and Technology

1 Oryong-dong Buk-gu Gwangju, 500-712, Korea

 

Phone: 010-5058-6832

E-mail: dkvmfhelxp34@gm.gist.ac.kr