Arun Goel

Application of support vector machines in scour prediction on grade-control structures

Research into the problem of predicting the maximum depth of scour on grade-control structures like sluice gates, weirs and check dams, etc., has been mainly of an experimental nature and several investigators have proposed a number of empirical relations for a particular situation. These traditional scour prediction equations, although offer some guidance on the likely magnitude of maximum scour depth, yet applicable to a limited range of the situations. It appears from the literature review that a regression mathematical model for predicting maximum depth of scour under all circumstances is not currently available. This paper explores the potential of support vector machines in modeling the scour from the available laboratory and field data obtained form the earlier published studies. To compare the results, a recently proposed empirical relation and a feed forward back propagation neural network model are also used in the present study. The outcome from the support vector machines-based modeling approach suggests a better performance in comparison to both the empirical relation and back propagation neural network approach with the laboratory data. The results also suggest an encouraging performance by the support vector machines learning technique in comparison to both empirical relation as well as neural network approach in scaling up the results from laboratory to field conditions for the purpose of scour prediction.

Stage and Discharge Forecasting by SVM and ANN Techniques

In this study, forecasting of stage and discharge was done in a time-series framework across three time horizons using three models: (i) persistence model, (ii) feed-forward neural network (FFNN) model, and (iii) support vector machine (SVM) model. For these models, lagged values of the time series constituted the set of input variables which had been selected by principal component analysis (PCA). Parameters of FFNN and SVM models were determined by sensitivity analysis. All the three models were evaluated using data from Mahanadi River, India, and their forecasting performance was then compared. It is shown that over a shorter forecasting horizon, it is difficult to outperform the persistence model. Moreover, results show that forecasting of stage and discharge over a longer time frame by the SVM model is more accurate than that by the other two models.

Modeling river discharge time series using support vector machine and artificial neural networks

Discharge time series were investigated using predictive models of support vector machine (SVM) and artificial neural network (ANN) and their performances were compared with two conventional models: rating curve (RC) and multiple linear regression (MLR) techniques. These models are evaluated using stage and discharge data from Big Cypress River, Texas, USA. Daily river stage–discharge data for the period of April 2010 to August 2013 were used for training and testing the above models and their results were compared using appropriate performance criteria. The evaluation of the results includes different performance measures, which indicate that SVM and ANN have an edge over the results by the conventional RC and MLR models. Notably, peak values predicted by SVM and ANN are more reliable than those by RC and MLR, although the performances of these conventional models are acceptable for a range of practical problems. The paper projects a critical view on inter-comparison studies by seeing through model selection approaches based on the common practice of the absolute best or even the best for the stated purpose towards uncertainty analysis.

Land Use Classification and Watershed Analysis of Assi River, Varanasi, Uttar Pradesh, India

Watershed analysis is essential for planning development activities or improving the features of a terrain. It gives an idea for various features like - aspect, elevation, slope, drainage, urban distribution, etc. in the area. This study is done either by field survey or with the help of various software tools. Varanasi has been selected as one of the cities to be developed into a smart city. But being one of the oldest cities of the world, a proper sustain planning is really essential to make this a reality. In the present study, river Assi (a tributary of Ganges) , geographically located between - 25°16’59.0” N and 83°00’35.3” E, in Varanasi district of Uttar Pradesh (India), has been considered. Continuous dumping of waste, heavy encroachments and improper planning has reduced this river into nothing but just a drain. Being a tributary of Ganges, all of this waste further reaches Ganges water, depleting its water quality too. Software’s like ArcGIS, ERDAS Imagine 2016 and SWAT has been used for the study of the watershed of this Assi River. The overall classification accuracy from the Land Use map for the 3rd order watershed has been computed as 89.32% with Kappa Coefficient being 0.7751. A digitized map of the watershed is prepared to compute the percentage of various features like - settlement, water bodies, cultivated land, etc. in the area of the watershed. Through SWAT, watershed has been divided into various sub-watersheds, which enables to in identification of key drains of the river. This study will thus not only help in identifying urban pattern of the area, but will also help in identifying key aspects that are to be answered in order to remediate Assi back to its river form through proper planning of development activities around Assi River without affecting its ecology.

Development of Stage-Discharge Relation Using Support Vector Machines

Establishment of rating curves are often required by the hydrologists for flow estimates in the streams, rivers etc. Measurement of discharge in a river is a timeconsuming, expensive, and difficult process. Furthermore, the conventional approach of regression analysis of the stage-discharge relation does not provide encouraging results especially during the floods. Therefore, present study is aimed at the application of support vector machines (SVM) based algorithm for modelling stagedischarge relation including the hysteresis effect. A data set of discharge-measuring station located on an Indian river has been used for analysis in the present study. A back propagation neural network model was also employed on the same data in order to compare the performance of the results based on support vector machines based modelling technique. The outcome of the study suggest that the support vector machines works quite well for both the data sets and produced promising results in comparison to the neural network technique. Finally, results also suggest the suitability of SVMs algorithm in predicting the looped rating curve having hysteresis effect as well.