하천수질 해석을 위한 2차원 이송-분산 모형의 종·횡혼합계수

초록

In this study, we collected and analyzed hydraulic and dispersion data measured in natural rivers. Then, using these data, we proposed new prediction equations for the longitudinal and transverse mixing coefficients of a two-dimensional advection-dispersion model. The result of the longitudinal mixing coefficient showed that the dimensionless values of the longitudinal mixing coefficient obtained from small and medium-sized rivers in Korea and the St. Clair River in the U.S. were found to be 10~400, which was much larger than the theoretical value of 5.93 proposed by Elder (1959). The result of analysis of the correlation between the longitudinal mixing coefficient and river hydraulic and topographic factors revealed that _ / has a positive correlation with / and / , but no significant correlation with / . The longitudinal mixing coefficient prediction equation derived in this study was found to have an excellent mean absolute percentage error of 13.8 %. In the case of the transverse mixing coefficient, _ / showed a linear relationship with / and / , expressed in logarithmic form, but no significant correlation was found with / . The results of comparing the transverse mixing coefficient prediction equation proposed in this study with existing empirical equations showed that the Bansal (1971) equation and the Deng et al. (2001) equation overpredict the measured values, and the Baek and Seo (2013) equation, Baek and Lee (2023) equation, and genetic programming based equation by Aghababaei et al. (2017) underpredict some measured values, while the proposed equation predicts values with high consistency with the measured values.

키워드