Local scour depth in a curved channel estimated by multivariable analysis method
Local scour depth in a curved channel estimated by multivariable analysis method / Song, Jai - Woo, Lee, Yong-Hwan, Kim, Du - Rae
p. 5-22; 26 cm
수록자료: Hongik journal of science and technology. Reasearch Institute for Science & Technology, Hong-Ik University. Vol.1(1997년 1월), p. 5-22 1<5 저자: Song, Jai-Woo, HongIk University, Professor, Civil Engineering 저자: Lee, Yong-Hwan, HongIk University, Course of Ph. D., Civil Engineering 저자: Kim, Du-Rae, HongIk University, Course of Master, Cilvil Engineering
Local scour around bridge piers is a complex problem that has been investigated extensively over the past few years. As a result, many equations exist for the prediction of scour depth. However, these equations mainly deal with straight channels. The experiments for maximum local scour depth around circular pier in three discharge conditions(Q = 20, 25, 30 1/sec) are conducted. According to the experimental results, the maximum velocity and the maximum transverse bed slope occur when the angle of bend is 150°. Based on measured values in a curved channel and by the nonlinear multiple regression analysis method considering the curvature ratio, the empirical equation predicting the maximum local scour depth in a curved channel is suggested for this channel. After residual analysis, the regression coefficients are determined in a curved channel by nonlinear multiple regression analysis to select the best regression model. Selected models are examined and the model's predicted values are compared with experimental data. Correlation analysis shows that the results from the CSU(Colorado State University) equation are much higher than those given by other equations. Generally, CSU equation yields values higher than the actual experimental data. The value predicted by the equation proposed in this study match closely with data from the laboratory flume experiments. Compared with previous predicted equations, the determination coefficient of equation presented in this study is higher than that of CSU equation. Compared with predicting equation presented by Choi in a curved channel, the equation proposed in this study superior to Choi equation.