Faculty Publications

Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736

Publications by NITK Faculty

Browse

Subject: search.filters.subject.Artificial neural network

Search Results

Now showing 1 - 10 of 21
  • No Thumbnail Available
    Item
    Estimation of Reservoir Storage Using Artificial Neural Network (ANN)
    (Springer Nature, 2018) Satish, P.; Ramesh, H.
    The rapid growth in population increases water demand thus resulting in scarcity of water which is due to improper management rather than lack of resources. Reservoir is the most important source for surface water. So, reservoir storage plays a crucial role in efficient reservoir management. Artificial neural network (ANN) is capable of simulating reservoir storage capacity. So, in the present work five different feed forward back propagation ANN models by varying number of hidden layer neurons were developed for estimation of Harangi reservoir storage, Karnataka, India. The first 2 years (2010–12) data was used for supervised training and remaining data (2013–14) was used in prediction. The predictive accuracy using the statistical parameters like correlation coefficient (R) and mean absolute percentage error (MAPE) were found within the acceptable limit. Result shows that, ANN model with five hidden neurons (i.e., network architecture of 6-5-1) is performing well compared to all other models for prediction of reservoir storage estimation. © Springer Nature Singapore Pte Ltd. 2019.
  • No Thumbnail Available
    Item
    Early cost estimation of highway projects in India using artificial neural network
    (Springer, 2019) Mahalakshmi, G.; C, C.
    The objective of this paper is to develop a model to estimate the construction cost of a highway at early stage (conceptual phase) of a project. Cost estimation at conceptual phase is a challenge as only limited information is known. As a result, wide cost variance is eminent at the completion of project. A neural network which can aid in cost estimation is developed. Parameters that can be obtained with least effort and highly influencing on cost are chosen. Developed neural network relates overall highway construction cost described in terms of materials, duration, topography, and prevailing soil conditions. Data of 52 projects were obtained from National Highway Authority of India (NHAI). The obtained results demonstrated that a multi perceptron network with backpropagation algorithm is capable of predicting construction cost of highway with reasonable accuracy. © Springer Nature Singapore Pte Ltd. 2019.
  • No Thumbnail Available
    Item
    Modeling of phenol degradation in spouted bed contactor using artificial neural network (ANN)
    (Walter de Gruyter GmbH, 2008) Dabhade, M.A.; Saidutta, M.B.; Murthy, D.V.R.
    Presence of phenol and phenolic compounds in various wastewaters and its harmful effects has led to the use of different treatment methods. Work on biological methods shows the use of different microorganisms and different bioreactors so as to improve the removal efficiency economically. The present work deals with the use of N. hydrocarbonoxydans (NCIM 2386), an actinomycetes, for the degradation of phenol. N. hydrocarbonoxydans was immobilized on GAC and used in a spouted bed contactor for effective contact of microorganisms and the substrate. The contactor performance was studied by varying flow rates, influent concentrations and the solids loading in the contactor. The effect of these variables on phenol degradation was investigated and modeling study was carried out using the artificial neural network (ANN). A feed forward neural network with back propagation was used for the model development. The experiments were planned as per the face centered cube design (FCCD) and used for training of the model, whereas data from four other experimental runs were used for testing and validation of the model. The network was optimized for the number of neurons based on the mean square error. The ANN model with three layers with three input neurons, eight neurons in hidden layers and one output neuron was found to predict effectively the effluent concentration for the given operating conditions in the spouted bed contactor. The mean square error was found to be 9.318e-12 for this ANN model. Also the experimental data was used to develop second order nonlinear empirical model obtained using multiple regression (MR) and the results compared with ANN using correlation coefficient (R2), average absolute error (AAE) and root mean square error (RMSE). Results show that R2, AAE and RMSE values of MR model were 0.9363, 2.085 % and 2.338 % respectively, while in case of ANN model these values were 0.9995, 0.59 % and 1.263 % respectively. This shows that ANN model prediction is better than multiple regression model prediction. Copyright © 2008 The Berkeley Electronic Press. All rights reserved.
  • No Thumbnail Available
    Item
    Raga classification for Carnatic music
    (Springer Verlag service@springer.de, 2015) Suma, S.M.; Koolagudi, S.G.
    In this work, an effort has been made to identify raga of given piece of Carnatic music. In the proposed method, direct raga classification without the use of note sequence has been performed using pitch as the primary feature. The primitive features that are extracted from the probability density function (pdf) of the pitch contour are used for classification. A feature vector of 36 dimension is obtained by extracting some parameters from the pdf. Since non-sequential features are extracted from the signal, artificial neural network (ANN) is used as a classifier. The database used for validating the system consists of 162 songs from 12 ragas. The average classification accuracy is found to be 89.5%. © Springer India 2015.
  • No Thumbnail Available
    Item
    Prediction of gallstone disease progression using modified cascade neural network
    (Springer Science and Business Media Deutschland GmbH info@springer-sbm.com, 2018) Thomas, L.; Manoj Kumar, M.V.; Annappa, B.; Arun, S.; Mubin, A.
    Prediction of disease severity is highly essential for understanding the progression of disease and initiating an early diagnosis, which is priceless in treatment planning. A Modified Cascade Neural Network (ModCNN) is proposed for stratification of the patients who may need Endoscopic Retrograde Cholangiopancreatography (ERCP). In this study, gallstone disease (GSD) whose prevalence is increasing in India is considered. A retrospective analysis of 100 patients was conducted and their case history was recorded along with the routine investigations. Using ModCNN, the associated risk factors were extracted for the prediction of disease progression toward severe complication. The proposed model outperformed showing better accuracy with an area under receiver operating characteristic curve (area under ROC curve) of 0.9793, 0.9643, 0.9869, and 0.9768 for choledocholithiasis, pancreatitis, cholecystitis, and cholangitis, respectively, when compared with Artificial Neural Network (ANN) showing an accuracy of 0.884. Hence, the proposed technique can be used to conduct a nonlinear statistical analysis for the better prediction of disease progression and assist in better treatment planning, avoiding future complications. © 2018, Springer Nature Singapore Pte Ltd.
  • No Thumbnail Available
    Item
    Inverse response surface method for structural reliability analysis
    (Springer Science and Business Media Deutschland GmbH, 2020) Nagesh, M.; Balu, A.S.
    Reliability-based design of complex structural systems is a computationally tedious task. In order to reduce the computational effort, approximation methods, such as classical response surface method, Kriging model and artificial neural network, can be adopted. Response surface model is a conventional method, where the limit state function is approximated using a suitable surrogate model. For the construction of response surface, variables of stochastic model should be known well in advance. However, the design parameters are unknown during initial stages of reliability-based design optimization (RBDO). For such structural design cases using RBDO, an adaptive inverse response surface procedure is proposed in this paper. The procedure is developed by coupling the adaptive response surface method with suitable experimental design (Halton low-discrepancy sequence sampling) for estimating reliability indicators and artificial neural network-based inverse reliability method for design optimization. The validity and accuracy of the proposed method are tested on example with explicit nonlinear limit state function. © Springer Nature Singapore Pte Ltd 2020.
  • No Thumbnail Available
    Item
    Heat transfer optimization using genetic algorithm and artificial neural network in a heat exchanger with partially filled different high porosity metal foam
    (Elsevier Ltd, 2022) Athith, T.S.; Trilok, G.; Jadhav, P.H.; Gnanasekaran, N.
    The metal foam is well known for its high surface area to volume ratio and thus used to transfer heat from the exhaust gas leaving the heat exchanger system. The present work deals with the numerical simulations of a heat exchanger partially filled with three different metal foams made up of Aluminum (Al), Copper (Cu) and Nickel (Ni) having two pore densities namely 20 PPI and 40 PPI, respectively. The hot gas is made to flow through the 8 mm channel in which metal foams are inserted and different heights of foams such as 2 mm, 4 mm, 6 mm and 8 mm are considered for the analysis. The purpose of this study is to optimize thermal performance by increasing heat transfer and decreasing pressure drop which is calculated from the simulations using a commercial software ANSYS FLUENT. In order to achieve this, a optimization technique called Non-dominated Sorting Genetic Algorithm (NSGA-II) is coded in MATLAB by making use of artificial neural network (ANN tool) as an interpolation tool to generate more data based on the already existing data. Finally, Pareto front is obtained for the optimized functional values of heat transfer and drop in pressure after running the code for NSGA-II. From the numerical simulations it is observed that there is 5.68 times enhancement in heat transfer rate when copper metal foam is used for higher inlet velocities, when compared with non-porous channel. From the optimization study, it is found that 50% filled metal foam porous channel is showing enhanced heat transfer rate with decreased pressure drop as depicted in the pareto optimal plot for copper and aluminium. © 2022 Elsevier Ltd.
  • No Thumbnail Available
    Item
    Performance Evaluation in 2D NoCs Using ANN
    (Springer Science and Business Media Deutschland GmbH, 2022) Kale, P.; Hazarika, P.; Jain, S.; Bhowmik, B.
    A network-on-chip (NoC) performance is traditionally evaluated using a cycle-accurate simulator. However, when the NoC size increases, the time required for providing the simulation results rises significantly. Therefore, such an issue must be overcome with an alternate approach. This paper proposes an artificial neural network (ANN)-based framework to predict the performance parameters for NoCs. The proposed framework is learned with the training dataset supplied by the BookSim simulator. Rigorous experiments are performed to measure multiple performance metrics at varying experimental setups. The results show that network latency is in the range of 31.74–80.70 cycles. Further, the switch power consumption is in the range of 0.05–12.41 μ W. Above all, the proposed performance evaluation scheme achieves the speedup of 277–2304 × with an accuracy of up to 93%. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
  • No Thumbnail Available
    Item
    Optimizing machining responses of homologous TiNiCu shape memory alloys using hybrid ANN-GA approach
    (Elsevier Ltd, 2022) Roy, A.; Sachin, B.; Raghavendra, T.; Rao, C.M.; Naik, G.M.; Soni, H.; Mashinini, P.M.; Narendranath, S.
    Fabrication of shape memory alloys using wire electro discharge machining (WEDM) has gained popularity over the last few years. Most widely used machining parameters of WEDM process are pulse on time (Øon), pulse off time (Øoff), servo voltage (σ) and wire feed (ω). WEDM responses like material removal rate (MR), surface roughness (SR), kerf width (KW) and recast layer thickness (LT) have been evaluated by researchers to determine machining characteristics and are also considered for this study. These machining responses determine the quality of machining and are majorly influenced by thermal conductivity and melting temperature of the WEDM workpiece. Actuation behavior of shape memory alloys is a function of phase transformation characteristics which in turn depends on elemental composition of the selected alloys. Therefore, dissimilar machining responses of Ti50Ni40Cu10 and Ti50Ni25Cu25 have been observed even though similar machining input values were used. This study utilized artificial neural network (ANN) mapping to establish WEDM response function – which was used as fitness function to perform multi objective optimization using genetic algorithm (GA). It was found that ANN successfully predicted machining responses of selected homologous alloys and GA helped in identifying suitable input parameter values to optimize machining responses. © 2022
  • No Thumbnail Available
    Item
    Predicting the Axial Load Carrying Capacity of Columns Reinforced with GFRP Rebars Using ANN Modelling
    (Springer Science and Business Media Deutschland GmbH, 2023) Sumesh Manohar, G.; Palanisamy, T.
    In recent years most of the concrete structures are getting exposed to environments that are resulting in the corrosion of steel. To eliminate this, studies have been carried out to replace steel in RCC by Glass Fiber Reinforced Polymer (GFRP) rebars. In this paper, several experimental results were considered and the impacts of substituting steel by GFRP rebars were studied. Parameters affecting the load-carrying capacity of columns reinforced with GFRP rebars were identified from various literature and a database has been created. Twelve such parameters describing the material property and geometric configuration are chosen as inputs and the axial load carrying capacity as an output. An ANN model is developed with optimized architecture for predicting the compressive strength of columns reinforced with GFRP rebars. The model is then trained, tested, and validated on this database. The accuracy of the ANN model is evaluated by various regression evaluation metrics such as MSE, RMSE and R2. Comparison with the existing empirical equations and code provisions showed that the ANN model outperformed all these models. For the purpose of determining the efficiency of ANN model, a subset of the experimental data collected from work done on GFRP reinforced columns is used. Sensitivity analysis is carried out and the results showed that the most important parameters for the estimation of the strength of GFRP reinforced columns are the geometrical dimensions of the column. The results obtained showed that the ANN model is in good agreement with the experimental results. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.