Modeling stream dissolved oxygen concentration using teaching–learning based optimization algorithm

摘要: The aim of this study is to investigate the applicability teaching–learning based optimization (TLBO) algorithm for first time in modeling stream dissolved oxygen (DO) prediction. input parameters selected from a surface water-quality including 20 indicators models are water pH, temperature, electrical conductivity, and hardness, which were measured semimonthly at six monitoring sites an untreated wastewater impacted urban during year, due their direct indirect effect on DO concentration. accuracy TLBO method compared with those artificial bee colony conventional regression analysis methods. These methods applied four different forms: quadratic, exponential, linear, power. There 144 data each indicator, 114 designated training rest testing patterns models. To evaluate performance models, five statistical indices, i.e., sum square error, root mean absolute average relative determination coefficient, used. quadratic form among all yielded better prediction, improvement nearly 20 %. It can be concluded that equations obtained by employing algorithms predict concentration successfully. Therefore, employment resources environment managers encouraged recommended future studies.