1. Ph.D Theses
Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/1/11
Browse
10 results
Search Results
Item Calibration of Vehicle and Driver Characteristics for Vissim Model, Ann-Based Sensitivity Analysis, Traffic Management, and Signal Design Using Ga for Mangalore City(National Institute of Technology Karnataka, Surathkal, 2022) Bandi, Marsh M.; George, VargheseThe field of traffic flow modeling has emerged as an important multi-disciplinary area with contributions from traffic-engineers, city-planners, mathematicians, and specialists in the field of computer sciences. Traffic engineers and planners constantly strive to alleviate problems that arise due to bottlenecks in traffic movement. One of the major challenges to traffic management lies in minimizing congestion and facilitating efficient traffic flow. The study of traffic congestion requires a proper understanding of the relationship between vehicle characteristics and driver characteristics to mimic existing traffic flows on urban streets.Simulation approaches permit traffic engineers of developing countries to evolve reliable models to investigate the influence of various factors related to roadways, and driver and vehicle characteristics on traffic flow on urban roads characterized by heterogeneous traffic conditions. These modeling techniques assist in gaining insight into the underlying relationships between the above factors involved. The primary scope of this study is focused on performing investigations using micro- simulation on understanding the traffic characteristics of Mangalore city in India, for heterogeneous traffic composition using VISSIM. A very important traffic circuit of the city connecting Hampankatta Circle, Navbharat Circle, PVS Circle, Bunts Hostel Circle, and Jyothi Circle, was considered for analysis, in addition to the nearby important locations such as Bendoorwell Junction, Balmatta Junction, and St. Theresa’s School Junction. In the initial phase of the study, the links and connectors representing the road network of the city were assigned in the VISSIM modeling environment on a template comprising a 1:5000 high- resolution base-map of the city overlaid with the layout of the roads and junctions using AutoCAD. Data on turning movements of traffic at various junctions was collected for 80 minutes during the peak-hour between 17:00 - 18.30 hours on Tuesday 10th March 2015 as part of this study. In this phase of the study, a number of simulation experiments were performed using the above data for default vehicle and driver characteristics, and the best random seed to be used for simulation was identified as 25, and 42 from among the random seeds from 1-50s tested. In the calibration exercise, the driver and vehicle parameters were fine-tuned in fourteen major stages to minimize errors between the observed data and the simulated results. 75% of the video-graphic data was used for performing testing and calibration, while the remaining 25% of the data was used for validation studies. An ANN-based sensitivity analysis was then performed to identify the relative importance of various vehicle and driver characteristics. A modified Garson’s approach was adopted in this study for the computation of relative sensitivity based on connection weights between the input layer, the three hidden layers, and the output layer for the optimized ANN configuration. Based on the results of the sensitivity analysis, the predictive capability of the simulation model was further enhanced by performing a multi-level extended calibration procedure that provided reliable results as per prescribed standards for traffic simulation. This finalized model was again validated successfully. The fully calibrated VISSIM model was the used inthe later phase of the study, to study the effect of implementing short-term strategies such as widening of existing road-widths, and long-term improvement strategies such as introduction of flyovers at selected critical locations in the city. Additionally, studies were performed using the genetic algorithm (GA) based approach in the design of traffic signal timings for streamlining traffic flows across four important junctions in the city. The objective function in the GA module was formulated based on the HCM average delay model (TRB 2000). The overall approach towards performing calibration studies evolved through the present study is expected to provide the basic framework for calibration and fine-tuning of vehicle and driver characteristics in the development of micro-simulation models. The findings of this study are expected to assist transport planners in developing innovative solutions to urban traffic management, analysis, design, and operation of vehicles on roadways.Item Inverse Estimation of Multi- Parameters Using Bayesian Framework Combined with Evolutionary Algorithms for Heat Transfer Problems(National Institute of Technology Karnataka, Surathkal, 2020) S, Vishweshwara P.; Gnanasekaran, N.; M, Arun.This thesis focuses on the estimation of unknown parameters using various inverse methods for the heat transfer problems. The first class of problem elaborately discusses about the estimation of interfacial heat transfer coefficients during the solidification of casting. To accomplish this, a prevalent one dimensional transient horizontal directional solidification of Sn-5%wtPb alloy with temperature dependent thermophysical properties and latent heat is considered to be the mathematical model/forward model and numerically solved using Explicit Finite Difference Method to obtain temperature distribution from the known boundary and initial conditions. The temperatures from the forward model is validated with the literature and an absolute error of 5% from the actual measurements was observed. In order to mimic the real time experiments, the temperatures are added with σ=0.01Tmax, σ=0.02Tmax and σ=0.03Tmax Gaussian white noise (simulated measurements) and compared with two different objective functions: (i) Least Squares and (ii) Bayesian Framework. Meantime, to expedite the solution of the inverse problem, the numerical model is then replaced with Artificial Neural Network (ANN), which acts as a fast forward model to estimate the unknown constants present in the correlation of interfacial heat transfer coefficient. A total of 473 data sets of inputs and corresponding outputs were used to create a trained artificial neural network which produced temperatures with an accuracy less than 0.1◦C temperature difference from the exact temperature data. Genetic Algorithm (GA) was implemented as an inverse method and it was found that ANN-GA-Bayesian framework was more effective compared to ordinary least squares for noise added data with an overall average error of 2%. Furthermore, an extended study on the advantage of Bayesian framework for the estimation of multi-parameters during Al-4.5wt%Cu alloy solidification is also discussed in detail. The main aim is to retrieve more information with less available simulated measurements. A sensitivity analysis is performed to understand the dependency of the unknown parameters like modeling error, latent heat and heat transfer coefficient parameters on the solution. It showed that the values of constants of the IHTC correlation and latent heat affect the temperature distribution in casting significantly. For iiithe solution of inverse estimation, the use of two different metaheuristic algorithms (i) Genetic Algorithm (GA) and (ii) Particle Swarm Optimization (PSO) is illustrated. A careful examination of the mentioned algorithms is performed to fix the algorithm parameters. The results revealed that PSO combined with Bayesian framework provides a better computational solution compared to GA-Bayesian with an overall absolute error less than 6%. Also, the study on the effect of multiple sensors revealed that using two sensor the average % error for the estimation of a ,b and latent heat was 0.247, 0.3 and 0.45 respectively and suggesting that two sensors were sufficient for the present analysis. The second class of problem is extended to retrieve the unknown heat flux and heat transfer coefficient for a 3-D steady state conjugate fin heat transfer problem. A mild steel fin with dimensions 150x250x6 mm3 is placed centrally on to an aluminium base of dimensions 150x250x8 mm3 and experiments are conducted for different heat flux values of 305, 544, 853 and 1232 W/m2 and corresponding temperature distribution along the vertical fin is recorded. Navier-Stokes equation is solved to obtain the necessary temperature distribution of the fin. Heat flux with the range between 305W/m2 and 3300 W/m2 and its corresponding temperature distribution of the fin is obtained using commercial software. A total of 24 Computational Fluid Dynamics (CFD) simulations are performed to create a neural network model that can surrogate the forward problem in order to expedite the computational process. The estimation of the heat flux and heat transfer coefficient using GA, PSO and PSO- Broyden Fletcher Goldfarb Shanno (BFGS) is carried out for both simulated and experimental data. A detailed comparison study on the effect of algorithm parameters on the solution is demonstrated in order to examine the performance of the algorithms. For simulated temperature measurements, all the mentioned algorithms proved to be effective but PSO-BFGS estimated the heat flux with an absolute % error of 0.86 and heat transfer coefficient with 0.105% for experimental temperatures. The results show that the PSO-BFGS method outperforms GA and PSO and is observed to be a formidable approach in the estimation of the unknown parametersItem Prediction of Local scour around bridge pier using Soft Computing Techniques(National Institute of Technology Karnataka, Surathkal, 2019) B. M, Sreedhara; Manu; Mandal, S.Bridges play an essential role in the society since they enable quick access across a river or any water body. Bridges facilitate transportation of goods and people and hence play a leading role in the development of a province. The safety of the bridge is the important factor with respect to scour failure which is the leading failure factor in river bridges. Scour is the removal of sediment near or around the structure which is located in the flowing water. There are different factors which affects scour mainly on the scour depth are flow depth, discharge, velocity, sediment size, porosity, pier shape and size etc. There are two types of scour conditions on which scour is classified and studied namely, clear water and live bed scour. The scour is the complex phenomenon and there is no common or general simple method to predict the scour depth around the bridge pier. There are several researchers who studied the scour mechanism using laboratory experiments. In the present days the artificial intelligence is the focal point for several researchers. Soft computing techniques, such as, Artificial Neural Network (ANN), Support Vector Machine (SVM), Adaptive Neuro-Fuzzy Inference System (ANFIS) and Particle Swarm Optimization (PSO) have been efficiently used for modeling scour related problems. The study used the data for developing the soft computing models, is obtained from a physical model study on scour depth around bridge pier, carried out by Goswami Pankaj in 2013 in a 2-D wave flume. The input parameters, namely, sediment size (d50), velocity (U), time (t) and sediment quantity (ppm) are used to predict the scour depth of different pier shapes such as circular, rectangular, round nosed and sharp nosed pier for both clear water and live bed scour condition. The complete original data is divided into training and testing. In the study, the soft computing techniques such as ANN, SVM, ANFIS, PSO-SVM and PSO-ANN are developed. The ANN model with feed-forward backpropagation network is developed with different hidden neurons. The RBF, Linear and Polynomial kernel functions are used in the SVM model. the ANFIS model is also developed with Trapezoidal, Gbell and Triangular membership function. The evolutionary optimization technique, particle swarm optimization is used to tune the SVM and ANN parameters to improve the efficiency of models prediction.ii The performance of individual and hybrid soft computing models are compared using statistical parameters such as, Correlation Coefficient (CC), Normalized Root Mean Square Error (NRMSE), Nash–Sutcliffe coefficient (NSE) and Normalized Mean Bias (NMB). Scatter plots are used to evaluate the accuracies of the models and box plots were used to analyze the spread or distribution of the data points estimated by the models. The validation of the developed models is done using the experimental values. The validation results shows that the proposed models are well correlated and in good agreement with experimental results. The hybrid models displayed a better performance compared to individual models. It is found that the hybrid PSO-SVM model is the best and efficient model in estimating the scour depth effectively around bridge pier for both live bed and clear water scour condition when compared to all the other models developed.Item Study of Geomorphology and Dynamics of Shoreline Associated with Mulky-Pavanje Rivermouth, Dakshina Kannada Coast, Karnataka, India(National Institute of Technology Karnataka, Surathkal, 2013) Nagaraj, Gumageri; Dwarakish, G. S.The current thesis considered Mulky-Pavanje rivermouth and associated shoreline of about 12km length, lies between 13000'00''-13006'00'' North Latitude and 74044'00''- 74050'00'' East Longitude of Dakshina Kannda coast, Karnataka, India for short-term (<10 years), medium-term (10–60 years) and long-term (>60 years) shoreline changes. Beach survey, beach width, wave climate (height, period and direction) and wind parameters (speed and direction) and sediment sampling are gathered from nine locations (BS 1 – BS 9) to represent total 12 km shoreline, during the period from September 2009 to December 2011 for short-term change analysis. Short-term change analysis indicated that net accretion on the beaches towards the south of the rivermouth (BS 1–BS 5), whereas the north of the rivermouth experienced net erosion (BS 6–BS 9). For medium-term shoreline change analysis, rainfall and river discharges are obtained from Indian Meteorological Department for the periods 1985- 2011 and 1985-1998 respectively. The monsoonal storm directly induces rivermouth morphology to vary (BS 5–BS 6), adjacent beaches to suffer from erosion (BS1–BS 4 and BS 7–BS 9) and also leads drastic changes in wave climate and freshwater flow. During monsoon and post-monsoon periods, the rivers Mulky (North) and Pavanje (South) overflow, discharge sizeable quantities of sediments into the sea, whereas during the pre-monsoon periods, seawater enters into the rivermouth area leads sediment deposition and distribution on either side of the rivermouth. However, the discharge of the Mulky river is approximately two times more than that of Pavanje river. Because of the more flow in the Mulky river, which runs across the northern part of the rivermouth, the shoreline in the vicinity of rivermouth is predominantly shifting towards south. Additionally long-term shoreline change analyses are made through multidated satellite imageries and topomaps for the period 1912–2009. The long-term shoreline change analyses depicts that northern spit and rivermouth are shifting towards south during the period 1912–2009 and also observed that fluctuation of accretion and erosion pattern on southern side of the shoreline is highly significant as compared with northern side. The Mulky-Pavanje rivermouth being highly complex and dynamic, but it provides wide scope for developmental activities around it. Therefore Land use/Land cover changes are attempted by considering recentix decade, i.e 1998–2009 with the help of topographical map and remote sensing data. Land use/Land cover change analysis indicated that, because of development of urbanization and industrialization around the rivermouth, the built-up area has been drastically increased, while the other coastal related geological features such as beach vegetation, mangroves and river sand are drastically reduced during the period 1998– 2009. In addition, Artificial Neural Network (ANN) technique is used to model the very important parameters of the coastal engineering such as wave height and littoral drift, which cause coastal erosion in the study area. The developed NARX and FFBP models are evaluated using error statistics. In both cases the NARX model performed better than FFBP and proved that wave height and littoral drift are the direct responsible factors to cause erosion in the Mulky-Pavanje rivermouth and associated shoreline.Item Groundwater Level Forecasting using Radial Basis Function and Generalized Regression Neural Networks(National Institute of Technology Karnataka, Surathkal, 2013) D, Sreenivasulu; Deka, Paresh ChandraForecasting of groundwater levels is very much useful for efficient planning in integrated management of groundwater and surface water resources in a basin. Accurate and reliable groundwater level forecasting models can help ensuring the sustainable use of a watershed’s aquifer for both urban and rural water supply. The present work investigates the potential of two Neural networks, such as Radial Basis Function Neural Networks (RBFNN) and Generalized Regression Neural Networks (GRNN) in comparison to regular ANN models like Feed Forward Back Propagation (FFBP) and Non-Linear Regression Model (NARX) for modeling in Ground water level (GWL) forecasting in a coastal aquifer at western Ghats of India. Total 24 wells (both shallow and deep) located within the study area (microwatershed of Pavanje river basin) were selected covering around 40sqkm. Here, two different dataset such as weekly Time series GWL and Meteorological variables those recorded during the study period (2004-2011) were used in the analysis. Various performance indices such as Root Mean Squared Error (RMSE), Coefficient of Correlation (CC) and Coefficient of Efficiency (CE) were used as evaluation criteria to assess the performance of the developed models. At the first stage, the potential and applicability of RBF for forecasting groundwater level are investigated. Weekly time series groundwater level data upto four lagged data has been used as various input scenario where predicted output are one and two week leadtime GWL. The analysis has been carried out separately for three representative open wells. For all the three well stations, higher accuracy and consistent forecasting performance for RBF network model was obtained compared to FFBP network model. After confirming the suitability of RBF in GWL forecasting and with better accuracy over FFBP, the work has been extended further to consolidate the applicability of RBF in multistep leadtime forecasting upto six week ahead. In this study, six representative wells are covered for development of RBF models for six different input combinations using lagged time series data. Outputs are the predicted GWL upto six week. RBF models are developed for every well station and results are compared with Non linear regression model (NARX). It has been observed that for allGroundwater level Forecasting using Radial Basis Function and Generalized Regression Neural Networks, Ph.D Thesis, 2012, NITK, Surathkal, India viii the six well station, the higher and consistent forecasting performance by RBF network model in multi step week lead which consolidates the forecasting capability of RBF. The NARX model result shows poor performance. In the third stage, to examine the potential and applicability of GRNN in GWL forecasting, various GRNN models has been developed by considering the advantage of S-summation and D-summation layers for different input combinations using time series data. Weekly time series groundwater level data upto four lagged data has been used as inputs where predicted outputs are one week leadtime GWL. The analysis has been carried out separately for three representative open wells. GRNN models were developed for every well and best model results were compared with best RBF and FFBP with LM training algorithm models. The RBF and GRNN models are almost performed similarly in GWL forecasting with higher accuracy in all the representative well station. The poor performance of FFBP-LM model is also satisfactory but found inferior than both GRNN and RBF. After confirming the potential and applicability of GRNN and RBF in time series GWL forecasting with similar capability, the robustness, adaptability and flexibility characteristics of these two techniques are further investigated for suitability with cause and effect relationship. Here various meteorological parameters are used as causable variable and the GWL is used as output effect .Only GRNN models are developed in the present study as RBF was found with similar predicting performance in previous studies. Five various input combinations are used to obtain best results as one step leadtime output for three representative wells. In this case also, GRNN model is predicting groundwater level with higher accuracy and with satisfactory results. The GRNN model performance is compared to general ANN (FFBP) model and found outperforming FFBP performance. The result of the study indicates the potential and suitability of RBFNN and GRNN modeling in GWL forecasting for multistep leadtime data. The performance of RBFNN and GRNN were found almost equally good. Although accuracy of forecasted GWL generally decreases with the increase of leadtime, the GWL forecast were obtained within acceptable accuracy for both the models.Item Effect Of Data Preprocessing On The Prediction Accuracy Of Artificial Neural Network Model in Hydrologic Time Series(National Institute of Technology Karnataka, Surathkal, 2013) Banhatti, Aniruddha Gopal; Deka, Paresh ChandraThe accurate prediction of hydrological behavior in both urban and rural watershed can provide valuable information for the urban planning, land use, design of civil projects and water resources management. Hydrology system is influenced by many factors such as weather, land cover, infiltration, evapotranspiration, so it includes a good deal of stochastic dependent component, multi-time scale and highly non-linear characteristics. Hydrologic time series are often non-linear and non- stationary. In spite of high flexibility of Artificial Neural Network (ANN) in modeling hydrologic time series, sometimes signals are highly non-stationary and exhibit seasonal irregularity. In such situation, ANN may not be able to cope with non-stationary data if pre-processing of input and/or output data is not performed. Pre-processing data refers to analyzing and transforming input and output variables in order to detect trends, minimize noise, underline important relationship and flatten the variables distribution in a time series. These analyses and transformations help the model learn relevant patterns. Pre-processing techniques, which facilitate stabilization of the mean and variance, and seasonality removal, are often applied to remove non- stationary aspect in data used to build soft computing models. In this study, different data pre-processing techniques are presented to deal with irregularity components that exist in a hydrologic time series data of the Brahmaputra basin within India at the Pandu gauging station near Guwahati city and Pancharatna gauging station further 150km downstream of Pandu by using daily time unit and their properties are evaluated by performing one step ahead flow forecasting using ANN. Three different preprocessed datasets are used for the analysis. Various ANN models are generated by varying network internal architecture with different input scenarios. The model results were evaluated by using Root Mean Square Error (RMSE)and Mean Absolute Percentage Error (MAPE) and found that Logarithmic based pre-processing techniques provide better forecasting performance among various pre-processing techniques. The results indicate that detecting non-stationary aspect and selecting an appropriate preprocessing technique is highly beneficial in improving the prediction performance of ANN model.Item Water Quality Assessment in Distribution System Using Artificial Intelligence(National Institute of Technology Karnataka, Surathkal, 2014) Krishnaji, Patki Vinayak; Shrihari, S.; Manu, B.In this study various artificial intelligence techniques have been compared for assessment and prediction of water quality in various zones of municipal distribution system using six physico-chemical characteristics viz. pH, alkalinity, hardness, dissolved oxygen (DO), total solids (TS) and most probable number (MPN). Fuzzy expert system, artificial neural network (ANN) and adaptive neuro fuzzy inference system (ANFIS) were used for the comparative study. The proposed expert system includes a fuzzy model consisting of IF-THEN rules to determine WQI based on water quality characteristics. The fuzzy models are developed using triangular and trapezoidal membership functions with centroid, bisector and mean of maxima (MOM) methods for defuzzification. In ANN method the cascade feed forward back propagation (CFBP) and feed forward back propagation (FFBP) algorithms were compared for prediction of water quality in the municipal distribution system. The comparative study was carried out by varying the number of neuron (1-10) in the hidden layer, by changing length of training dataset and by changing transfer function. ANFIS models are developed by using triangular, trapezoidal, bell and Gaussian membership function. Further, these artificial intelligence techniques are compared with multiple linear regression technique, which is the commonly used statistical technique for modelling water quality variables. The study revealed that artificial neural network (ANN) outperforms other modelling techniques and is a robust tool for understanding the poorly defined relations between water quality variables and water quality index (WQI) in municipal distribution system. This tool could be of great help to the distribution system operator and manager to find change in WQI with changes in water quality varibles.Item Damage Level Prediction of NonReshaped Berm Breakwater using Soft Computing Techniques(National Institute of Technology Karnataka, Surathkal, 2014) N, Harish.; Rao, Subba; Mandal, SukomalTranquility condition inside the port and harbor has to be maintained for loading cargo and passengers. In order to maintain calm condition inside the port and harbor, breakwater has to be constructed to dissipate wave energy that is coming inside. The alignment of the breakwater must be carefully considered after examining the predominant direction of approach of waves and winds, degree of protection required, magnitude and direction of littoral drift and the possible effect of these breakwaters on the shoreline. In general these studies are invariably conducted in a physical model test where various alternatives are studied and the final selection will be based on performance consistent with cost. Considering the coastal boundary and depth variation, field analysis of wave structure interaction, determination of stability and damage level of berm breakwater structure is difficult. Mathematical modeling of these complex interactions is difficult while physical modeling will be costly and time consuming. Hence one has to depend on physical model studies which are expensive and time consuming. Soft computing techniques, such as, Artificial Neural Network (ANN), Support Vector Machine (SVM),Adaptive Neuro-Fuzzy Inference System (ANFIS) and Particle Swarm Optimization (PSO) have been efficiently proposed as a powerful tool for modeling and predictions in coastal/ocean engineering problems. For developing soft computing models in prediction of damage level of non-reshaped berm breakwater, data set are obtained from experimental damage level of non-reshaped berm breakwater using regular wave flume at Marine Structure Laboratory, National Institute of Technology, Karnataka, Surathkal, Mangalore, India. These data sets are divided into two groups, one for training and the other for testing. The input parameters that influence the damage level (S) of nonreshaped berm breakwater, such as, relative wave steepness (H/L0), surf similarity (ζ), slope angle (cotα) relative berm position by water depth (hB/d), relative armour stone weight (W50/W50max), relative berm width (B/ L0) and relative berm location (hB/L0) are considered in developing soft computing models for prediction damage level. The ANN model is developed for the prediction of damage level of non-reshaped berm breakwater. Two network models, ANN1 and ANN2 are constructed based on the parameters which influence the damage level of non-reshaped berm breakwater. The seven input parameters that are initially considered for ANN1 model are (H/L0), (ζ), (cotii α), (hB/d), (W50/W50max), (B/ L0) and (hB/L0). The ANN1 model is studied with different algorithm namely, Scaled Conjugate Gradient (SCG), Gradient Descent with Adaptive learning (GDA) and Levenberg-Marquardt Algorithm (LMA) with five numbers of hidden layer nodes and a constant 300 epochs. LMA showed good performance than the other algorithms. Also, influence of input parameters is evaluated using Principal Component Analysis (PCA). From PCA study, it is observed that cotα is the least influencing parameter on damage level. Based on the PCA study, least influencing parameter is discarded and ANN2 model is developed with remaining six input parameters. Training and testing of the ANN2 network models are carried out with LMA for different hidden layer nodes and epochs. The ANN2 with LMA 6-5-1 with 300 epochs gave good results. It is observed that the correlation of about 88% between predicted and observed damage level values by the ANN2 network models and measured values are in good agreement Furthermore, to improve the result of prediction of damage level of non-reshaped berm breakwater, SVM model was developed. This technique works on structural risk minimization principle that has greater generalization ability and is superior to the empirical risk minimization principle as adopted in conventional neural network models. This model was developed based on statistical learning theory. The basic idea of SVM is to map the original data x into a feature space with high dimensionality through a nonlinear mapping function and construct an optimal hyper-plane in new space. SVM models were constructed using different kernel functions. In order to study the performance of each kernel in predicting damage level of non-reshaped berm breakwater, SVM is trained by applying these kernel functions. Performance of SVM is based on the best setting of SVM and kernel parameters. Correlation Coefficient (CC) of SVM (polynomial) model (CC Train = 0.908 and CC Test = 0.888) is considerably better than other SVM models. To avoid over-fitting or under-fitting of the SVM model due to the improper selection of SVM and kernel parameters and also the performance of SVM, hybrid particle swarm optimization tuned support vector machine regression (PSO-SVM) model is developed to predict damage level of non-reshaped berm breakwater. The performance of the PSOSVM models in the prediction of damage level is compared with the measured values using statistical measures, such as, CC, Root mean Square Error (RMSE) and Scatteriii Index (SI). PSO-SVM model with polynomial kernel function gives realistic prediction when compared with the observed values (CC Train = 0.932, CC Test = 0.921). It is observed that the PSO-SVM models yield higher CCs as compared to that of SVM models. However, it is noticed that ANN model in isolation cannot capture all data patterns easily. Adaptive Neuro-Fuzzy Inference System (ANFIS) uses hybrid learning algorithm, which is more effective than the pure gradient decent approach used in ANN. ANFIS models were developed with different membership namely Triangular-shaped built-in membership function (TRIMF), Trapezoidal-shaped built-in membership function (TRAPMF), Generalized bell-shaped built-in membership function (GBELLMF), and Gaussian curve built-in membership function (GAUSSMF) to predict damage level of non-reshaped berm breakwater. The performance of the ANFIS models in the prediction of damage level is compared with the measured values using statistical measures, such as, CC, RMSE and SI. ANFIS model with GAUSSMF gave realistic prediction when compared with the observed values (CC Train = 0.997, CC Test = 0.938). It is observed that the ANFIS models yield higher CCs as compared to that of ANN models. The different soft computing models namely, ANN, SVM, PSO-SVM and ANFIS results are compared in terms of CC, RMSE, SI and computational time. The hybrid models in both (ANFIS and PSO-SVM) cases showed better results compared to individual models (ANN and SVM). When the hybrid models are compared, ANFIS model gives higher CC and lower RMSE. But considering computational time, ANFIS has taken more time than PSO-SVM model. Hence PSO-SVM is computationally efficient as compared to ANFIS. ANFIS and PSO-SVM models perform better and similar to observed values. Hence, ANFIS or PSO-SVM can replace the ANN, SVM for damage level prediction of nonreshaped berm breakwater. ANFIS or PSO-SVM can be utilized to provide a fast and reliable solution in prediction of the damage level prediction of non-reshaped berm breakwater, thereby making ANFIS or PSO-SVM as an alternate approach to map the wave structure interactions of berm breakwater.Item Modeling reference evapotranspiration using hybrid artificial intelligence techniques in arid and semi-arid regions of India(National Institute of Technology Karnataka, Surathkal, 2017) Patil, Amit Prakash; Deka, Paresh ChandraEvapotranspiration (ET) plays an important role in efficient crop water management. Accurate estimation of ET is a challenging task in developing countries like India, where the availability of meteorological data is often minimal. This study makes an attempt to evaluate the potential and applicability of hybrid Wavelet-Artificial Intelligence (AI) models for estimating reference crop evapotranspiration (ETo) in arid and semi-arid regions of India. The hybrid models were developed by using wavelet decomposed subseries of meteorological variables as inputs to the ANN, ANFIS and LS-SVM models. Performance of the proposed hybrid models was then compared to the classical AI models. The study used forty year weekly dataset from Jodhpur and Pali (arid region) weather station. Also, daily data for six years were obtained from Hyderabad and Kurnool weather station (semi-arid region). In absence of lysimeter data, ETo values are calculated by FAO-56PM equation. Prior to the development of models, factor analysis test was employed to identify the input combination that may yield more efficient model under limited data scenario. Additionally, the effectiveness of using ETo data from another station in the same climatic region (extrinsic data) was also evaluated. It is expected that the proposed hybrid models together with extrinsic input variables would provide efficient ETo estimation models. The performance of hybrid and classical AI models were compared using RMSE, NSE and threshold statistics. Scatter plots were used to evaluate the accuracies of the models and box plots were used to analyze the spread of the data points estimated by the models. The results show that the proposed AI models worked better at estimating weekly ETo in arid regions compared to estimating daily ETo in semi-arid regions. The hybrid AI models displayed a better performance compared to the classical AI models at all the stations. It was found that hybrid W-LSSVM was the best model for estimating ETo in both arid and semi-arid region. Further, it was observed that the use of extrinsic inputs delivered good results only in arid regions. It was also observed that in semi-arid regions, use of wavelet decomposed extrinsic data deteriorated the performance of some hybrid models.Item Inverse Techniques for the Estimation of Multiple Parameters Using Steady State Heat Transfer Experiments(National Institute of Technology Karnataka, Surathkal, 2018) M. K, Harsha Kumar; N, GnanasekaranThe aim of the present research work is to estimate the unknown parameters by using the information obtained from in-house steady state heat transfer experiments and to employ stochastic inverse techniques. With the advent of latest technologies in the field of advance computing, conjugate heat transfer problems that are highly complex can easily be solved to obtain temperature distributions. In the present work, suitable mathematical models are proposed as forward models for a class of conjugate heat transfer problem. The first problem solved was a conjugate heat transfer from a mild steel fin. The numerical model is developed using ANSYS FLUENT with an extended model which facilitates natural convection heat transfer. Based on the experimental temperatures and with accompanying mathematical model, heat flux is estimated using Genetic Algorithm as inverse method. To accelerate the inverse estimation, Genetic algorithm is assisted with the Levenberg- Marquardt method for the estimation of the heat flux, thus making the whole process as hybrid estimation. In the second problem, 3-D conjugate fin model is proposed for the estimation of heat flux and heat transfer coefficient using Artificial Neural Network (ANN) method. The novelty of the work is to inject the experimental temperature methodologically in to the forward model which is trained by Neural network thereby the forward model is driven by experimental data and to accomplish the task of parameter estimation, ANN is used as inverse method that leads to a non-iterative solution. The concept of a priori information is then introduced for the simultaneous estimation of heat flux and heat transfer coefficient using experimental data. This was accomplished using Bayesian framework along with Markov Chain Monte Carlo (MCMC) method to condition the posterior probability density function. A powerful Metropolis-Hastings algorithm is exploited in order to attain stable Markov chains during the process of inverse estimation. Finally, this was followed by estimation of heat generation and heat transfer coefficient from a Teflon cylinder within the Bayesian framework.