Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
64 results
Search Results
Item Developing regression models for predicting pan evaporation from climatic data - A comparison of multiple least-squares, principal components, and partial least-squares approaches(2007) Kovoor, G.M.; Nandagiri, L.Regression models for predicting daily pan evaporation depths from climatic data were developed using three multivariate approaches: multiple least-squares regression (MLR), principal components regression (PCR), and partial least-squares (PLS) regression. The objective was to compare the prediction accuracies of regression models developed by these three approaches using historical climatic datasets of four Indian sites that are located in distinctly different climatic regimes. In all cases (three approaches applied to four climatic datasets), regression models were developed using a part of the data and subsequently validated with the remaining data. Results indicated that although performances of the regression models varied from one climate to another, more or less similar prediction accuracies were obtained by all three approaches, and it was difficult to identify the best approach based on performance statistics. However, the final forms of the regression models developed by the three approaches differed substantially from one another. In all cases, the models derived using PLS contained the smallest number of predictor variables; between two to three out of a possible maximum of six predictor variables. The MLR approach yielded models with three to six predictor variables, and PCR models included all six predictor variables. This implies that the PLS regression models are the most parsimonious in terms of input data required for estimating epan from climate variables, and yet yield predictions that are almost as accurate as the more data-intensive MLR and PCR models. © 2007 ASCE.Item Prediction of water quality indices by regression analysis and artificial neural networks(2008) Rene, E.R.; Saidutta, M.B.The quality of wastewater generated in any process industry is generally indicated by performance indices namely BOD, COD and TOC, expressed in mg/L. The use of TOC as an analytical parameter has become more cornmon in recent years especially for the treatment of industrial wastewater. In this study, several empirical relationships were established between BOD and COD with TOC using regression analysis, so that TOC can be used to estimate the accompanying BOD or COD. A new, the use of Artificial Neural Networks has been explored in this study to predict the concentrations of BOD and COD, well in advance using some easily measurable water quality indices. The total data points obtained from a refinery wastewater (143) were divided into a training set consisting of 103 data points, while the remaining 40 were used as the test data. A total of 12 different models (Al-A12) were tested using different combinations of network architecture. These models were evaluated using the % Average Relative Error values of the test set. It was observed that three models gave accurate and reliable results, indicating the versatility of the developed models.Item Artificial neural networks model for the prediction of steady state phenol biodegradation in a pulsed plate bioreactor(2008) Shetty K, K.V.; Nandennavar, S.; Srinikethan, G.Background: A recent innovation in fixed film bioreactors is the pulsed plate bioreactor (PPBR) with immobilized cells. The successful development of a theoretical model for this reactor relies on the knowledge of several parameters, which may vary with the process conditions. It may also be a time-consuming and costly task because of their nonlinear nature. Artificial neural networks (ANN) offer the potential of a generic approach to the modeling of nonlinear systems. Results: A feedforward ANN based model for the prediction of steady state percentage degradation of phenol in a PPBR by immobilized cells of Nocardia hydrocarbonoxydans (NCIM 2386) during continuous biodegradation has been developed to correlate the steady state percentage degradation with the flow rate, influent phenol concentration and vibrational velocity (amplitude x frequency). The model used two hidden layers and 53 parameters (weights and biases). The network model was then compared with a Multiple Regression Analysis (MRA) model, derived from the same training data. Further these two models were used to predict the percentage degradation of phenol for blind test data. Conclusions: The performance of the ANN model was superior to that of the MRA model and was found to be an efficient data-driven tool to predict the performance of a PPBR for phenol biodegradation. © 2008 Society of Chemical Industry.Item Neuro-fuzzy based approach for wave transmission prediction of horizontally interlaced multilayer moored floating pipe breakwater(2011) Patil, S.G.; Mandal, S.; Hegde, A.V.; Alavandar, S.The ocean wave system in nature is very complicated and physical model studies on floating breakwaters are expensive and time consuming. Till now, there has not been available a simple mathematical model to predict the wave transmission through floating breakwaters by considering all the boundary conditions. This is due to complexity and vagueness associated with many of the governing variables and their effects on the performance of breakwater. In the present paper, Adaptive Neuro-Fuzzy Inference System (ANFIS), an implementation of a representative fuzzy inference system using a back-propagation neural network-like structure, with limited mathematical representation of the system, is developed. An ANFIS is trained on the data set obtained from experimental wave transmission of horizontally interlaced multilayer moored floating pipe breakwater using regular wave flume at Marine Structure Laboratory, National Institute of Technology Karnataka, Surathkal, India. Computer simulations conducted on this data shows the effectiveness of the approach in terms of statistical measures, such as correlation coefficient, root-mean-square error and scatter index. Influence of input parameters is assessed using the principal component analysis. Also results of ANFIS models are compared with that of artificial neural network models. © 2010 Elsevier Ltd. All rights reserved.Item Prediction of uniaxial compressive strength, tensile strength and porosity of sedimentary rocks using sound level produced during rotary drilling(2011) Rajesh Kumar, B.R.; Vardhan, H.; Govindaraj, M.The main purpose of the study is to develop a general prediction model and to investigate the relationships between sound level produced during drilling and physical properties such as uniaxial compressive strength, tensile strength and percentage porosity of sedimentary rocks. The results were evaluated using the multiple regression analysis taking into account the interaction effects of various predictor variables. Predictor variables selected for the multiple regression model are drill bit diameter, drill bit speed, penetration rate and equivalent sound level produced during rotary drilling (Leq). The constructed models were checked using various prediction performance indices. Consequently, it is possible to say that the constructed models can be used for practical purposes. © Springer-Verlag 2011.Item Land use scenario analysis and prediction of runoff using SCS-CN method: A case study from the Gudgudi tank, Haveri district, Karnataka, India(2011) Bhagwat, N.B.; Shetty, A.; Hegde, V.S.Runoff from the Gudgudi tank catchment (209 ha) near Hangal in the Northern Karnataka is estimated employing Soil Conservation Services(SCS) model based on the hydrological data and land use/ land cover data. Rainfall measured for 2006 using a tipping bucket indicated annual rainfall of 887.7mm in the tank catchment. Textural characteristics of the soil indicate sandy-clayey type which corresponds to hydrological soil group "C and D". Average Soil infiltration rate of 0.18 cm/hour for the forest-land and 0.21 cm/hour for agriculture land has been observed. Weighted curve number is arrived based on the antecedent moisture conditions, and runoff is estimated for the existing land-use. Areastorage curve is constructed using the tank bed contours. Considering the hypothetical changes in the agriculture and forest area coverage, optimum conditions for maximizing the runoff and storage in the tank is arrived. The analysis suggests land use pattern of 15% of forest cover and 85% of agriculture land coverage in this region provide maximum runoff and storage in the tank for sustainable development. © 2011 CAFET-INNOVA TECHNICAL SOCIETY.Item Experimental investigation and artificial neural network-based modeling of batch reduction of hexavalent chromium by immobilized cells of newly isolated strain of chromium-resistant bacteria(2012) Shetty K, K.V.; Namitha, L.; Rao, S.N.; Narayani, M.The batch bioreduction of Cr(VI) by the cells of newly isolated chromium-resistant Acinetobacter sp. bacteria, immobilized on glass beads and Ca-alginate beads, was investigated. The rate of reduction and percentage reduction of Cr(VI) decrease with the increase in initial Cr(VI) concentration, indicating the inhibitory effect of Cr(VI). Efficiency of bioreduction can be improved by increasing the bioparticle loading or the initial biomass loading. Glass bioparticles have shown better performance as compared to Ca-alginate bioparticles in terms of batch Cr(VI) reduction achieved and the rate of reduction. Glass beads may be considered as better cell carrier particles for immobilization as compared to Ca-alginate beads. Around 90% reduction of 80 ppm Cr(VI) could be achieved after 24 h with initial biomass loading of 14.6 mg on glass beads. Artificial neural networkbased models are developed for prediction of batch Cr(VI) bioreduction using the cells immobilized on glass and Ca-alginate beads. © Springer Science+Business Media B.V. 2011.Item Development of prediction models for hydrodynamic performance of semicircular breakwater(2012) Aggarwal, A.; Gope, V.K.; Managiri, S.S.; Hegde, A.V.Breakwaters are structures built to protect harbors, shore areas, basins, and other areas from the fury of sea waves. They create calm waters and provide for the safe mooring and handling of ships, as well as protection to harbor facilities. The main function of a breakwater is the formation of an artificial harbor. Of late, certain new types of breakwaters have been constructed to cater to the tranquility requirements of managing marine traffic in ports. The semicircular breakwater (SBW) is one such new type of breakwater. The semicircular breakwater possesses a round top and, thus, offers more stability against the action of waves. It is expected that the SBW will be well suited as an offshore breakwater designed to protect beaches from coastal erosion. A number of experiments were conducted on scaled-down physical models of SBW for different values of parameters such as wave height H, wave period T, spacing of perforations on the seaside, etc. (radius of breakwater and diameter of perforations were kept constant), and data were collected. The paper presents the prediction models/equations for hydrodynamic performance characteristics such as reflection coefficient and relative wave runup, using the data obtained by a regression approach in MATLAB.Item Prediction of penetration rate and sound level produced during percussive drilling using regression and artificial neural network(2012) Kivade, S.B.; Murthy, C.S.N.; Vardhan, H.The main objective of this investigation is to develop a general prediction model and to study the effect of predictor variables such as uniaxial compressive strength, air pressure and thrust on penetration rate and sound level produced during percussive drilling of rocks. The experiment was carried out using three levels Box-Behnken design with full replication in 15 trials. Modeling was done using artificial neural network (ANN) and multipleregression analysis (MRA). These techniques can be utilized for the prediction of process parameters. Comparison of artificial neural network and multiple linear regression models was made and found that error rate was smaller in ANN than that predicted by MRA in terms of sound level and penetration rate. © 2012 CAFET-INNOVA TECHNICAL SOCIETY.Item Regression analysis and ANN models to predict rock properties from sound levels produced during drilling(Elsevier Ltd, 2013) Rajesh Kumar, B.; Vardhan, H.; Govindaraj, M.; Vijay, G.S.This study aims to predict rock properties using soft computing techniques such as multiple regression, artificial neural network (MLP and RBF) models, taking drill bit speed, penetration rate, drill bit diameter and equivalent sound level produced during drilling as the input parameters. A database of 448 cases were tested for determination of uniaxial compressive strength (UCS), Schmidt rebound number (SRN), dry density (?), P-wave velocity (Vp), tensile strength (TS), modulus of elasticity (E) and percentage porosity (n) and the prediction capabilities of the models were then analyzed. Results from the analysis demonstrate that neural network approach is efficient when compared to statistical analysis in predicting rock properties from the sound level produced during drilling. © 2012 Elsevier Ltd.