Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
13 results
Search Results
Item In the present study, various ocean wave parameters are estimated from theoretical Pierson-Moskowitz spectra as well as measured ocean wave spectra using backpropagation neural networks (BNN). Ocean wave parameters estimation by BNN shows that the correlations are very close to one. This substantiates the use of neural networks (NN). For Indian coast, Scott spectra are used as it reasonably represents the measured spectra. The correlations of NN and Scott spectra are also compared. Once the network is trained, the ocean wave parameters can be estimated for unknown measured spectra, whereas significant wave height and spectral peak period are required to first generate the Scott spectra and then estimate other ocean wave parameters. © 2005 Elsevier Ltd. All rights reserved.(Ocean wave parameters estimation using backpropagation neural networks) Mandal, S.; Rao, S.; Raju, D.H.2005Item 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 A differential evolution based neural network approach to nonlinear system identification(2011) Subudhi, B.; Jena, D.This paper addresses the effectiveness of soft computing approaches such as evolutionary computation (EC) and neural network (NN) to system identification of nonlinear systems. In this work, two evolutionary computing approaches namely differential evolution (DE) and opposition based differential evolution (ODE) combined with Levenberg Marquardt algorithm have been considered for training the feed-forward neural network applied for nonlinear system identification. Results obtained envisage that the proposed combined opposition based differential evolution neural network (ODE-NN) approach to identification of nonlinear system exhibits better model identification accuracy compared to differential evolution neural network (DE-NN) approach. The above method is finally tested on a one degree of freedom (1DOF) highly nonlinear twin rotor multi-input-multi-output system (TRMS) to verify the identification performance. © 2010 Elsevier B.V. All rights reserved.Item Artificial neural network based modeling to evaluate methane yield from biogas in a laboratory-scale anaerobic bioreactor(Elsevier Ltd, 2016) Nair, V.V.; Dhar, H.; Kumar, S.; Thalla, A.K.; Mukherjee, S.; Wong, J.W.C.The performance of a laboratory-scale anaerobic bioreactor was investigated in the present study to determine methane (CH4) content in biogas yield from digestion of organic fraction of municipal solid waste (OFMSW). OFMSW consists of food waste, vegetable waste and yard trimming. An organic loading between 40 and 120 kg VS/m3 was applied in different runs of the bioreactor. The study was aimed to focus on the effects of various factors, such as pH, moisture content (MC), total volatile solids (TVS), volatile fatty acids (VFAs), and CH4 fraction on biogas production. OFMSW witnessed high CH4 yield as 346.65 L CH4/kg VS added. A target of 60–70% of CH4 fraction in biogas was set as an optimized condition. The experimental results were statistically optimized by application of ANN model using free forward back propagation in MATLAB environment. © 2016 Elsevier LtdItem A non-invasive approach to investigation of ventricular blood pressure using cardiac sound features(IOP Publishing Ltd, 2017) Tang, H.; Zhang, J.; Chen, H.; Mondal, A.; Park, Y.Heart sounds (HSs) are produced by the interaction of the heart valves, great vessels, and heart wall with blood flow. Previous researchers have demonstrated that blood pressure can be predicted by exploring the features of cardiac sounds. These features include the amplitude of the HSs, the ratio of the amplitude, the systolic time interval, and the spectrum of the HSs. A single feature or combinations of several features have been used for prediction of blood pressure with moderate accuracy. Experiments were conducted with three beagles under various levels of blood pressure induced by different doses of epinephrine. The HSs, blood pressure in the left ventricle and electrocardiograph signals were simultaneously recorded. A total of 31 records (18 262 cardiac beats) were collected. In this paper, 91 features in various domains are extracted and their linear correlations with the measured blood pressures are examined. These features are divided into four groups and applied individually at the input of a neural network to predict the left ventricular blood pressure (LVBP). The analysis shows that non-spectral features can track changes of the LVBP with lower standard deviation. Consequently, the non-spectral feature set gives the best prediction accuracy. The average correlation coefficient between the measured and the predicted blood pressure is 0.92 and the mean absolute error is 6.86 mmHg, even when the systolic blood pressure varies in the large range from 90 mmHg to 282 mmHg. Hence, systolic blood pressure can be accurately predicted even when using fewer HS features. This technique can be used as an alternative to real-time blood pressure monitoring and it has promising applications in home health care environments. © 2017 Institute of Physics and Engineering in Medicine.Item Back propagation genetic and recurrent neural network applications in modelling and analysis of squeeze casting process(Elsevier Ltd, 2017) Gowdru Chandrashekarappa, M.; Shettigar, A.K.; Krishna, P.; Parappagoudar, M.B.Today, in competitive manufacturing environment reducing casting defects with improved mechanical properties is of industrial relevance. This led the present work to deal with developing the input-output relationship in squeeze casting process utilizing the neural network based forward and reverse mapping. Forward mapping is aimed to predict the casting quality (such as density, hardness and secondary dendrite arm spacing) for the known combination of casting variables (that is, squeeze pressure, pressure duration, die and pouring temperature). Conversely, attempt is also made to determine the appropriate set of casting variables for the required casting quality (that is, reverse mapping). Forward and reverse mapping tasks are carried out utilizing back propagation, recurrent and genetic algorithm tuned neural networks. Parameter study has been conducted to adjust and optimize the neural network parameters utilizing the batch mode of training. Since, batch mode of training requires huge data, the training data is generated artificially using response equations. Furthermore, neural network prediction performances are compared among themselves (reverse mapping) and with those of statistical regression models (forward mapping) with the help of test cases. The results shown all developed neural network models in both forward and reverse mappings are capable of making effective predictions. The results obtained will help the foundry personnel to automate and précised control of squeeze casting process. © 2017Item ANN model for prediction of bit–rock interface temperature during rotary drilling of limestone using embedded thermocouple technique(Springer Science and Business Media B.V., 2020) Vijay Kumar, V.K.; Kunar, B.M.; Murthy, C.S.N.In the present work, an artificial neural network (ANN) model has been developed to predict the bit–rock interface temperature using a newly fabricated grounded K-type thermocouple (range 0–1250 °C) during rotary drilling in a CNC vertical machining center. The data have been taken from experimental observation using an embedded thermocouple technique in the laboratory at room temperature (28 °C) using a masonry drill bit. The observations were made using four different operational conditions, namely drill bit diameter (6, 8, 10, 12 and 16 mm), spindle speed (250, 300, 350, 400 and 450 rpm), rate of penetration (2, 4, 6, 8 and 10 mm min?1) and depth (6, 14, 22 and 30 mm). The ANN has been developed based on the multi layer perceptron neural network (MLPNN) with four different input parameters. A Levenberg–Marquardt (LM) algorithm with feed-forward and backward propagation has been used in this model. The predicted value of the bit–rock interface temperature with the highest R2 value provides a satisfactory result with the experimental data. The training value of RMSE is 1.2127, MAPE is 0.0196 and R2 is 0.9960, while the testing value of RMSE is 1.2770, MAPE is 0.0170 and R2 is 0.9978. The ANN model shows that the proposed MLPNN model successfully predicts the bit–rock interface temperature during the rotary drilling of limestone. © 2019, Akadémiai Kiadó, Budapest, Hungary.Item Artificial bee colony, genetic, back propagation and recurrent neural networks for developing intelligent system of turning process(Springer Nature, 2020) Shettigar, A.K.; Gowdru Chandrashekarappa, G.C.M.; Ganesh, G.R.; Vundavilli, P.R.; Parappagoudar, M.B.Intelligent manufacturing requires significant technological interventions to interface manufacturing processes with computational tools in order to dynamically mold the systems. In this era of the 4th industrial revolution, Artificial neural network (ANNs) is a modern tool equipped with a better learning capability (based on the past experience or history data) and assists in intelligent manufacturing. This research paper reports on ANNs based intelligent modelling of a turning process. The central composite design is used as a data-driven modelling tool and huge input–output is generated to train the neural networks. ANNs are trained with the data collected from the physics-based models by using back-propagation algorithm (BP), genetic algorithm (GA), artificial bee colony (ABC), and BP algorithm trained with self-feedback loop. The ANNs are trained and developed as both forward and reverse mapping models. Forward modelling aims at predicting a set of machining quality characteristics (i.e. surface roughness, cylindricity error, circularity error, and material removal rate) for the known combinations of cutting parameters (i.e. cutting speed, feed rate, depth of cut, and nose radius). Reverse modelling aims at predicting the cutting parameters for the desired machining quality characteristics. The parametric study has been conducted for all the developed neural networks (BPNN, GA-NN, RNN, ABC-NN) to optimize neural network parameters. The performance of neural network models has been tested with the help of ten test cases. The network predicted results are found in-line with the experimental values for both forward and reverse models. The neural network models namely, RNN and ABC-NN have shown better performance in forward and reverse modelling. The forward modelling results could help any novice user for off-line monitoring, that could predict the output without conducting the actual experiments. Reverse modelling prediction would help to dynamically adjust the cutting parameters in CNC machine to obtain the desired machining quality characteristics. © 2020, Springer Nature Switzerland AG.Item Application ANN Tool for Validation of LHD Machine Performance Characteristics(Springer, 2020) Balaraju, B.; Raj, G.R.; Murthy, C.S.Survival of industries has become more critical in the present global competitive business environment unless they produce their projected production levels. The accomplishment of this can be possible only by maintaining the men and machinery in an efficient and effective manner. Hence, it is more essential to estimate the performance of utilized equipment for reaching/achieving future goals. The present study focuses on the estimation of underground mining machinery such as the load–haul–dump machine performance characteristics using ‘Isograph Reliability Workbench 13.0’ software. The allocation of best-fit/goodness-of-fit distribution was made by utilizing the Kolmogorov–Smirnov test (K–S) test. The parameters were recorded based on the best-fitted results using the maximum likelihood estimate test. Further, a feed-forward-back-propagation artificial neural network (ANN) tool has been used to develop the models of reliability, availability and preventive maintenance time intervals. The number of neurons was selected with the Levenberg–Marquardt learning algorithm in the hidden layer as the optimal value. The output responses were predicted corresponding to the optimal values. Further, an attempt has been made to validate the computed results with ANN predicted responses. The recommendations are suggested to the industry based on the results for the improvement of system performance. © 2020, The Institution of Engineers (India).