Journal
KSCE JOURNAL OF CIVIL ENGINEERING
Volume 22, Issue 7, Pages 2206-2213Publisher
KOREAN SOCIETY OF CIVIL ENGINEERS-KSCE
DOI: 10.1007/s12205-017-1703-6
Keywords
ant colony for continuous domain; differential evolution. Electrical Conductivity (EC); Genetic Algorithm (GA); Sodium Absorption Ratio (SAR); Total Hardness (TH)
Categories
Ask authors/readers for more resources
Water quality management and control has high importance in planning and developing of water resources. This study investigated application of Genetic Algorithm (GA), Ant Colony Optimization for Continuous Domains (ACO(R)) and Differential Evolution (DE) in improving the performance of adaptive neuro-fuzzy inference system (ANFIS), for evaluating the quality parameters of Gorganroud River water, such as Electrical Conductivity (EC), Sodium Absorption Ratio (SAR) and Total Hardness (TH). Accordingly, initially most suitable inputs were estimated for every model using sensitivity analysis and then all of the quality parameters were predicted using mentioned models. Investigations showed that for predicting EC and TH in test stage, ANFIS-DE with R-2 values of 0.98 and 0.97, respectively and RMSE values of 73.03 and 49.55 and also MAPE values of 5.16 and 9.55, respectively were the most appropriate models. Also, ANFIS-DE and ANFIS-GA models had the best performance in prediction of SAR (R-2 = 0.95, 0.91; RMSE = 0.43, 0.37 and MAPE = 13.43, 13.72) in test stage. It is noteworthy that ANFIS showed the best performance in prediction of all mentioned water quality parameters in training stage. The results indicated the ability of mentioned algorithms in improving the accuracy of ANFIS for predicting the quality parameters of river water.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available