In river research, forecasting flow velocity accurately in vegetated channels is a significant challenge. The forecasting performance of various independent and hybrid machine learning (ML) models are thus quantified for the first time in this work. Utilizing flow velocity measurements in both natural and laboratory flume experiments, we assess the efficacy of four distinct standalone machine learning techniques-Kstar, M5P, reduced error pruning tree (REPT) and random forest (RF) models. In addition, we also test for eight types of hybrid ML algorithms trained with an Additive Regression (AR) and Bagging (BA) (AR-Kstar, AR-M5P, AR-REPT, AR-RF, BA-Kstar, BA-M5P, BA-REPT and BA-RF). Findings from a comparison of their predictive capabilities, along with a sensitivity analysis of the influencing factors, indicated: (1) Vegetation height emerged as the most sensitive parameter for determining the flow velocity; (2) all ML models displayed outperforming empirical equations; (3) nearly all ML algorithms worked optimal when the model was built using all of the input parameters. Overall, the findings showed that hybrid ML algorithms outperform regular ML algorithms and empirical equations at forecasting flow velocity. AR-M5P (R
Abstract We present Python Statistical Analysis of Turbulence (P-SAT), a lightweight, Python framework that can automate the process of parsing, filtering, computation of various turbulent statistics, spectra computation for steady flows. P-SAT framework is capable to work with single as well as on batch inputs. The framework quickly filters the raw velocity data using various methods like velocity correlation, signal-to-noise ratio (SNR), and acceleration thresholding method in order to de-spike the velocity signal of steady flows. It is flexible enough to provide default threshold values in methods like correlation, SNR, acceleration thresholding and also provide the end user with an option to provide a user defined value. The framework generates a .csv file at the end of the execution, which contains various turbulent parameters mentioned earlier. The P-SAT framework can handle velocity time series of steady flows as well as unsteady flows. The P-SAT framework is capable to obtain mean velocities from instantaneous velocities of unsteady flows by using Fourier-component based averaging method. Since P-SAT framework is developed using Python, it can be deployed and executed across the widely used operating systems. The GitHub link for the P-SAT framework is: https://github.com/mayank265/flume.git .
Sediment deposition impacts the hydraulic capacity of a channel in urban drainage and sewer systems. To reduce the impact of this continuous deposition of sediment particles, sewer systems are typically designed with a self-cleansing mechanism to keep the bottom of the channel clean from sedimentation. Therefore, accurate prediction of the particle Froude number (Fr) is important in designing sewer systems. This study used five data sets available in the literature, comprising wide ranges of the volumetric sediment concentration (Cv), dimensionless grain size of particles (Dgr), sediment median size (d), hydraulic radius (R), pipe friction factor (λ) for the condition of nondeposition with deposited bed. Five different input variable combinations were considered for the prediction of Fr. Four boosting machine-learning models, i.e., AdaboostRegressor, GradientBoostingRegressor, CatboostRegressor, and LightGBMRegressor, were developed, and the results obtained were compared with the existing empirical equations as well as state-of-the-art approaches proposed in the literature. To evaluate the proposed models, several performance metrics were used, such as index of agreement (Id), mean absolute error (MAE), root-mean-square error (RMSE), R2, and adjusted R2. AdaboostRegressor (Id=0.981, MAE=0.483, RMSE=0.591, R2=0.940, and adjusted R2=0.937) provided better results, followed by GradientBoostingRegressor, CatboostRegressor, and LightGBMRegressor. The boosting techniques used in this study performed better than multigene genetic programming, gene expression programming, multilayer perceptron (MLP), and the empirical equations proposed in the literature, indicating superior performance.
This paper presents a simplified technique to simulate strong ground motion from a finite source of an earthquake. The simplified technique is based on modifications made in the semi empirical technique given by Midorikawa [1993] and later modified by Joshi and Midorikawa [2004]. Modifications in this technique have been made to consider the effect of radiation pattern and seismic moment of the target earthquake. The coastal region of Sumatra Island was struck by a great earthquake of magnitude 9.0 (M w ) on 26th December, 2004. This earthquake is known for its release of high amount of energy and the devastating Tsunami. This earthquake was recorded at several broadband stations including a nearest broadband station located in Indonesia. The source of this earthquake is modeled by a finite rectangular rupture plane. Various locations of nucleation point and different values of rupture velocity have been tested before finalizing the rupture responsible for this earthquake. Iterative modeling and comparison of simulated and observed record due to final model suggests that the rupture initiated at the western end of the rupture plane at a depth of 38 km and started propagating in all direction with a rupture velocity of 3.0 km/s. The final model has been used to simulate record at MDRS and VISK stations located at the coastal region of India and simulated records are compared with observed records at these stations. The comparisons confirm the suitability of final model for predicting strong ground motion and the efficacy of the approach in modeling great earthquake. Strong ground motion has been simulated for the Sumatra earthquake of 26th December, 2004 at various hypothetical stations surrounding the final model of rupture plane. The distribution of peak ground acceleration in the near source region has been computed from simulated record at these stations. The isoacceleration contours shows that high peak acceleration zones of the order of > 2 g are observed in the source zone of this earthquake which gradually decreases with distance. Using the parameters of final model of the Sumatra earthquake a great hypothetical earthquake at northern segment of Andaman ridge has been modeled and records have been simulated at Port Blair (POR) station located in the Andaman Island, India. The simulated records shows that peak ground acceleration of the order of 1.4 g can be observed at POR station due to a hypothetical earthquake in the Andaman Island suggesting high seismic hazard in this region.
'%3e%3ccircle r='20' fill='%23008'/%3e%3ccircle r='17.5' fill='%23fff'/%3e%3ccircle r='3.5' fill='%23008'/%3e%3cg id='d'%3e%3cg id='c'%3e%3cg id='b'%3e%3cg id='a' fill='%23008'%3e%3ccircle r='.875' transform='rotate(7.5 -8.75 133.5)'/%3e%3cpath d='M0 17.5.6 7 0 2l-.6 5L0 17.5z'/%3e%3c/g%3e%3cuse xlink:href='%23a' transform='rotate(15)'/%3e%3c/g%3e%3cuse xlink:href='%23b' transform='rotate(30)'/%3e%3c/g%3e%3cuse xlink:href='%23c' transform='rotate(60)'/%3e%3c/g%3e%3cuse xlink:href='%23d' transform='rotate(120)'/%3e%3cuse xlink:href='%23d' transform='rotate(-120)'/%3e%3c/g%3e%3c/svg%3e)
