Modeling total resistance and form resistance of movable bed channels via experimental data and a kernel-based approach

Abstract

An accurate prediction of roughness coefficient in alluvial channels is of substantial importance for river management. In this study, the total and form resistance in alluvial channels with dune bedform were assessed using experimental data. First, the data of experiments carried out at the Hydraulic Laboratory of University of Tabriz was used to investigate the impact of hydraulic and sediment parameters on roughness coefficient. Then, these data were combined with other laboratory data, and the total and bedform resistance were modeled via a Gaussian Process Regression (GPR) approach. For models, developing different input combinations were considered based on flow and sediment characteristics. The obtained results from the experiments showed that the Reynolds number has a better correlation with flow resistance in comparison with other hydraulic parameters. It was found that the roughness variations due to bedform are almost between 40 and 80% of the total roughness coefficient. Also, the obtained results proved the capability of the GPR method in the modeling process. It was found that the model which took the advantages of both flow and sediment characteristics performed better compared to the other models. The sensitivity analysis results showed that the Reynolds number has the most significant impact in the prediction process.

Estimation of Manning Roughness Coefficient in Alluvial Rivers with Bed Forms Using Soft Computing Models

“…A comparison of the statistical error indices (R 2 and RMSE) for the models developed in the present study and the models developed by previous scholars (Roushangar et al 2018;Roushangar et al 2017;Saghebian et al 2020) indicated that the soft computing models developed in the present study had significantly higher accuracy.…”

“…To compare the efficiency of the soft computing models used in this study in estimating the Comparing the performance of soft computing models used in this study with models used in previous studies Roushangar et al (2018) showed that the models developed by them (i.e., FFNN, RBNN, and ANFIS) to estimate the Manning roughness coefficient in rivers with a dune bed form had R 2 ≤ 0.9 in the verification stage. Saghebian et al (2020) estimated the Manning roughness coefficient in dune and ripple rivers using multilayer perceptron-firefly algorithm (MLP-FFA) and feed-forward neural network (FFNN) models. They reported R 2 = 0.56 and RMSE = 0.0034 for the mentioned model in the best scenario.…”

Insights into the prediction capability of roughness coefficient in current ripple bedforms under varied hydraulic conditions

“…Figure 4 depicts the scatter plots between the observed data and model results. The results of previous works (Roushangar 2017(Roushangar , 2018Saghebian 2020) show that the best results for dune-bed channel studies for prediction of Manning roughness coefficient were obtained from GEP (R ¼ 0.866, NSE ¼ 0.742, and RMSE ¼ 0.0035), least squares SVM (R ¼ 0.839, NSE ¼ 0.705, and RMSE ¼ 0.0036), and GPR (R ¼ 0.784 and NSE ¼ 0.715), respectively. Hence, it seems that hydraulic conditions governing ripple bedforms provide better predictive ability for machine learning approaches in comparison to channels with dune bedforms.…”

“…Roushangar et al (2018aRoushangar et al ( , 2018b applied extreme learning machine (ELM) in order to find the nonlinear interaction among different input variables for the prediction of coefficient of friction of overland flows. More recently, Saghebian et al (2020) presented the applicability of Gaussian process regression (GPR) for the prediction of total and bedform resistance of dune bed channels.…”

A new formula for predicting movable bed roughness coefficient in the Middle Yangtze River

“…Movable bed roughness formulas were also proposed through theoretical or empirical methods using flume or field measurements, which can be divided into two types of formulas. The first kind of formula develops empirical or semi-empirical relationships based on the hypotheses that hydraulic radius or energy gradient can be linearly divided into two components covering grain roughness and bedform drag (Einstein and Barbarosssan, 1952; Engelund, 1966; Niazkar, 2020; Saghebian et al, 2020; Yang et al, 2005). It is difficult to investigate the bedform configurations in natural rivers, with the data of bedform geometry and motion being limited.…”