Hydrology & Meteorology
Total Page:16
File Type:pdf, Size:1020Kb
Vikas Kumar Vidyarthi et al., Hydrol Current Res 2014, 5:4 http://dx.doi.org/10.4172/2157-7587.S1.013 3rd International Conference on Hydrology & Meteorology September 15-16, 2014 Hyderabad International Convention Centre, India Investigation of sensitivity of popular training methods to initial weights in ANN rainfall-runoff modeling Vikas Kumar Vidyarthi and Ashu Jain Indian Institute of Technology Kanpur, India unoff estimation is a key input in any water resource management activity. It is generally estimated by developing rainfall- Rrunoff (RR) models. There are many techniques employed for RR modeling and artificial neural network (ANN) is one of the popular methods among them. The gradient descent (GD) and Levenberg-Marquardt (LM) optimization methods are commonly adopted algorithms for the training of ANN models. It has been reported that the performance of these algorithms is always sensitive to their initial weights. In this paper, the sensitivity of these two training algorithms to initial weights in the performance of ANN-RR model was investigated. The best ANN architecture was determined using a trial and error procedure in which the number of hidden neurons was varied from 1 to 20 and the architecture giving best performance in terms of certain error statistics was selected as the best. Each of the twenty architectures was trained using BPA and LMA and the best architecture was selected, named ANN-BPA and ANN-LMA, respectively. Then, these best ANN architectures were trained on ten different set of initial weights using both BPA and LMA. The performance of the best ANN model trained by BPA and LMA on different initial weights was then compared using standard error statistical measures. The daily rainfall, runoff data derived from Bird creek basin, Oklahoma, USA have been employed to develop all the models included here. The input variables were selected on the basis of correlation analysis. The performance evaluation statistics such as average absolute relative error (AARE), Pearson’s correlation coefficient (R) and threshold statistics (TS) were used for comparing all the models developed using both the optimization algorithm here. Based on the results obtained in this study, it has been found that the LMA trained ANN model performed better than the BPA trained ANN model. Further, the LMA trained ANN model is found to be more robust than the BPA trained ANN model as the ten different set of initial weights result into final solution similar to each other in case of the LMA trained ANN models. Biography Vikas Kumar Vidyarthi is a PhD student in the Department of Civil Engineering, IIT Kanpur. [email protected] Hydrol Current Res 2014 Volume 5, Issue 4 ISSN: 2157-7587, HYCR an open access journal Hydrology-2014 September 15-16, 2014 Page 72 Ila Dashora et al., Hydrol Current Res 2014, 5:4 http://dx.doi.org/10.4172/2157-7587.S1.013 3rd International Conference on Hydrology & Meteorology September 15-16, 2014 Hyderabad International Convention Centre, India Comparative study of ARIMA, Thomas Fiering and ANN models for streamflow generation of intermittent river for Narmada River basin Ila Dashora, S K Singal and D K Srivastav Indian Institute of Technology Roorkee, India ynthetic generation of streamflow data facilitates the planning and operation of water resource projects. Short term synthetic Sstreamflow generation helps to operate multipurpose water resource projects whereas long-term forecasting facilitates flood control operations. Significance of streamflow generation for intermittent river increases many fold so that available water can be use yearlong. In the present study monthly streamflow data for intermittent river Goi of Narmada river basin is been used. Herein the performance of stochastic streamflow generation models - ARIMA (p, d, q) and Thomas-Fiering model are being compared with Artificial Neural Network approach. The study reveals that ANN performs better than stochastic models. The neural network application in proportion of 70:15:15 for training, validation and testing respectively performs best among other ratios. Stochastic model ARIMA (2, 1, 2) has more reliability over Thomas-Fiering model. The performance is measured on the basis of RMSE and coefficient of determination. This forecasting is helpful for small dam construction so that water can be harnessed for multipurpose utility. [email protected] Hydrol Current Res 2014 Volume 5, Issue 4 ISSN: 2157-7587, HYCR an open access journal Hydrology-2014 September 15-16, 2014 Page 73 Prasoon Kumar Singh et al., Hydrol Current Res 2014, 5:4 http://dx.doi.org/10.4172/2157-7587.S1.013 3rd International Conference on Hydrology & Meteorology September 15-16, 2014 Hyderabad International Convention Centre, India Water contamination index: Important tools for groundwater quality assessment Prasoon Kumar Singh and Poornima Verma Indian School of Mines, India ater is the essential part for the development of any area; in which groundwater is one of the most valuable resource Wand one of the basic needs for human being. The inadequate availability of surface water makes people dependent on ground water resources to fulfill their needs. Ground water is the major source of water supply for drinking purposes in most parts of India. It is also our most important sources for irrigation. But unfortunately, due to the rapid increase in population, intense urbanization and large extent of industrialization, ground water quality is being increasingly threatened by agricultural chemicals and disposal of urban and industrial wastes. It has been estimated that once pollution enters the subsurface environment, it may remain concealed for many years, becoming dispersed over wide areas of ground water aquifer and rendering ground water supplies unsuitable for consumption and other uses. The quality of water contamination index evaluation attempt single value which decreases the big quantity of parameters and represents data in a simple way. Water Contamination Index (Cd) is an effective tool for evaluating and mapping the degree of groundwater contamination. It is the sum of the individual factors of single component that exceeds the maximum permissible concentration of water quality parameters. These tools have been used for the classification of the contamination in terms of Low, Medium, or High grade. These methods provide a suitable technique to evaluate the actual and potential groundwater contamination of any area for environmental scientist and decision-makers. The present paper highlights in brief about the different contamination index methods for evaluation of groundwater quality assessment. [email protected] , [email protected] Hydrol Current Res 2014 Volume 5, Issue 4 ISSN: 2157-7587, HYCR an open access journal Hydrology-2014 September 15-16, 2014 Page 74 Ashwani Kumar Tiwari et al., Hydrol Current Res 2014, 5:4 http://dx.doi.org/10.4172/2157-7587.S1.013 3rd International Conference on Hydrology & Meteorology September 15-16, 2014 Hyderabad International Convention Centre, India Quality assessment of mine water in the West Bokaro coalfield area, India Ashwani Kumar Tiwari, Prasoon Kumar Singh and Mukesh Kumar Mahato Indian School of Mines, India est Bokaro coalfield plays an important role in Indian coal production. Coal is exploited by both opencast as well as Wunderground mining methods and during this process, a huge quantity of water is discharged from coal mines to the natural drainage to facilitate safe mining. The discharged mine water varies greatly in the concentration of contaminants present, and in some cases it may even meet the drinking water specification. A geochemical study of mine water in the West Bokaro coalfield has been undertaken to assess its quality and suitability for domestic, industrial and irrigation uses. For this purpose, fifteen mine water samples collected from different mining areas of West Bokaro coalfield were analysed for pH, 2+ 2+ + + - l- electrical conductivity (EC), major cations (Ca , Mg , Na , K ), anions (F , C , HCO3-, SO42- and NO3-) and trace metals. pH of the analyzed water samples varied from 6.6 to 8.3 and the average pH was found to be 7.7 indicating mildly acidic to 2- 2+ 2+ slightly alkaline nature. SO4 and HCO3- are dominant in the anion and Ca and Mg in the cation chemistry of mine water. 2- The drinking water quality assessment indicates that number of mine water samples have high TDS, total hardness and SO4 , concentrations. Concentrations of some trace metals (i.e., Fe, Mn, Ni) were found to be above the levels recommended for drinking water. However, the mine water is good to permissible quality and suitable for irrigation, except at some sites, where higher salinity and Mg-ratio restrict its suitability for irrigation at some sites. Biography Ashwani Kumar Tiwari is the youngest researcher in the field of Environmental Hydrogeochemistry, Water Resources Management and GIS. He is pursuing PhD research work entitled “GIS Based Aquifer Vulnerability Assessment and Qualitative Analysis of Water Resources in a Coal Mining Area, Jharkhand” at Indian School of Mines (ISM) Dhanbad, Jharkhand. He has done MSc and MPhil in Environmental Science with first class. He has worked as a Research Fellow about one & half years at Central Institute of Mining and Fuel Research (CIMFR), Dhanbad, Jharkhand, India. He has published eight research papers in peer-reviewed international and national journals. He has also contributed four chapters in a research book and edited the proceeding of