Showing papers by "Kongu Engineering College published in 2013"

Characterization of natural fiber and composites – A review:

Abstract: The natural fiber-reinforced polymer composite materials offered extensive range of properties which are suitable for large number of engineering application. The natural fibers have been abundantl...

Characterization of new cellulose sansevieria ehrenbergii fibers for polymer composites

Abstract: Natural cellulose fibers were newly identified from the sources of sansevieria ehrenbergii plant. These fibers were extracted using the mechanical decortication process. The hierarchical cell structure of the plant and fibers was analyzed using scanning electron microscope, optical microscope, Fourier transforms infrared, and X-ray diffraction. The density and diameter of the fibers were found to be approximately 0.887 g/cm3 and 10–250 μm, respectively. The various chemical compositions were analyzed and compared with other natural fibers. The thermal stability of the fiber was examined through thermogravimetric analysis/differential thermogravimetric analysis (DTG). The maximum peak temperature was obtained at 333.02 °C in DTG curve. The raw fibers exhibited a tensile strength of 50–585 MPa, an elongation at break of 2.8–21.7%, a Young’s modulus of 2.5–7.5 GPa, and a corrected compliances Young’s modulus of 2.5–7.8 GPa.

Artificial neural network and response surface methodology modeling in mass transfer parameters predictions during osmotic dehydration of Carica papaya L.

Abstract: In this study, a comparative approach was made between artificial neural network (ANN) and response surface methodology (RSM) to predict the mass transfer parameters of osmotic dehydration of papaya. The effects of process variables such as temperature, osmotic solution concentration and agitation speed on water loss, weight reduction, and solid gain during osmotic dehydration were investigated using a three-level three-factor Box-Behnken experimental design. Same design was utilized to train a feed-forward multilayered perceptron (MLP) ANN with back-propagation algorithm. The predictive capabilities of the two methodologies were compared in terms of root mean square error (RMSE), mean absolute error (MAE), standard error of prediction (SEP), model predictive error (MPE), chi square statistic (χ2), and coefficient of determination (R2) based on the validation data set. The results showed that properly trained ANN model is found to be more accurate in prediction as compared to RSM model.

Sentiment analysis and classification based on textual reviews

Abstract: Mining is used to help people to extract valuable information from large amount of data. Sentiment analysis focuses on the analysis and understanding of the emotions from the text patterns. It identifies the opinion or attitude that a person has towards a topic or an object and it seeks to identify the viewpoint underlying a text span. Sentiment analysis is useful in social media monitoring to automatically characterize the overall feeling or mood of consumers as reflected in social media toward a specific brand or company and determine whether they are viewed positively or negatively on the web. This new form of analysis has been widely adopted in customer relation management especially in the context of complaint management. For automating the task of classifying a single topic textual review, document-level sentiment classification is used for expressing a positive or negative sentiment. So analyzing sentiment using Multi-theme document is very difficult and the accuracy in the classification is less. The document level classification approximately classifies the sentiment using Bag of words in Support Vector Machine (SVM) algorithm. In proposed work, a new algorithm called Sentiment Fuzzy Classification algorithm with parts of speech tags is used to improve the classification accuracy on the benchmark dataset of Movies reviews dataset.

Mechanical properties of randomly oriented snake grass fiber with banana and coir fiber-reinforced hybrid composites:

Abstract: Composite materials are extensively used in the structural applications due to its enhanced load-bearing capabilities. Mostly, the mechanical properties of the natural fiber-reinforced hybrid compo...

Design and optimization of machining fixture layout using ANN and DOE

Abstract: In machining fixtures, minimizing workpiece deformation due to clamping and cutting forces is essential to maintain the machining accuracy. This can be achieved by selecting the optimal location of fixturing elements such as locators and clamps. Many researches in the past decades described more efficient algorithms for fixture layout optimization. In this paper, artificial neural networks (ANN)-based algorithm with design of experiments (DOE) is proposed to design an optimum fixture layout in order to reduce the maximum elastic deformation of the workpiece caused by the clamping and machining forces acting on the workpiece while machining. Finite element method (FEM) is used to find out the maximum deformation of the workpiece for various fixture layouts. ANN is used as an optimization tool to find the optimal location of the locators and clamps. To train the ANN, sufficient sets of input and output are fed to the ANN system. The input includes the position of the locators and clamps. The output includes the maximum deformation of the workpiece for the corresponding fixture layout under the machining condition. In the testing phase, the ANN results are compared with the FEM results. After the testing process, the trained ANN is used to predict the maximum deformation for the possible fixture layouts. DOE is introduced as another optimization tool to find the solution region for all design variables to minimum deformation of the work piece. The maximum deformations of all possible fixture layouts within the solution region are predicted by ANN. Finally, the layout which shows the minimum deformation is selected as optimal fixture layout.

Effect of Particle Size on Tribological Behavior of Rice Husk Ash–Reinforced Aluminum Alloy (AlSi10Mg) Matrix Composites

Abstract: Rice husk ash of three different particle size ranges (50–75, 75–100 and 100–150 μm) a 3, 6, 9, and 12% by weight is reinforced with an aluminum alloy (AlSi10Mg) using the liquid metallurgy method. The dry sliding wear behavior of the composites in the cast conditions is examined using the pin-on-disc tribotesting machine for three different loads (20, 30, and 40 N) with three different sliding velocities (2, 3, and 4 m/s). The results reveal that the composite reinforced with the coarse rice husk ash particles exhibits superior wear resistance compared to the fine rice husk ash particles. The wear rate of the composite decreased with an increase in the weight percentage of rice husk ash particles for all size ranges. Finally, the wear mechanism was investigated with the worn surface using a scanning electron microscope.

An enhanced bacterial foraging algorithm approach for optimal power flow problem including FACTS devices considering system loadability.

Abstract: Obtaining optimal power flow solution is a strenuous task for any power system engineer. The inclusion of FACTS devices in the power system network adds to its complexity. The dual objective of OPF with fuel cost minimization along with FACTS device location for IEEE 30 bus is considered and solved using proposed Enhanced Bacterial Foraging algorithm (EBFA). The conventional Bacterial Foraging Algorithm (BFA) has the difficulty of optimal parameter selection. Hence, in this paper, BFA is enhanced by including Nelder-Mead (NM) algorithm for better performance. A MATLAB code for EBFA is developed and the problem of optimal power flow with inclusion of FACTS devices is solved. After several run with different initial values, it is found that the inclusion of FACTS devices such as SVC and TCSC in the network reduces the generation cost along with increased voltage stability limits. It is also observed that, the proposed algorithm requires lesser computational time compared to earlier proposed algorithms.

Mechanical and machinability behaviors of woven coir fiber-reinforced polyester composite

Abstract: The research on coir-polyester composites initiated the interest in the development of woven coir fiber-reinforced polyester composites. The mechanical properties of woven coir-polyester composites were evaluated as per ASTM standards and the machinability behavior was studied by conducting drilling tests in this investigation. The woven coir-polyester composites exhibited the average values of tensile, flexural and impact strength of 19.9 MPa, 31.3 MPa and 49.9 kJ/m2 respectively. The effect of NaOH treatment on the improvement of mechanical properties of woven coir-polyester composites were studied in this investigation. The 40 % increase of tensile strength, 42 % increase of flexural strength and 20 % increase of impact strength were achieved by treated woven coir fiber-reinforced polyester composites. The regression models for predicting thrust force, torque and tool wear in drilling of woven coir-polyester composites were developed and the effect of drilling parameters were analyzed.

Preparation of nanostructured TiO2 photoelectrode for flexible dye-sensitized solar cell applications

Abstract: Nanocrystalline titanium dioxide (TiO2) thin film was successfully prepared by simple electrodeposition method from alkaline aqueous solution containing potassium titanium oxalate and hydroxylamine. Surface characterization of the electrodeposited films indicates the formation of crystalline TiO2. The dye solar cell constructed from dye-modified electrodeposited TiO2 film achieved an overall light-to-electricity conversion efficiency of 2.1 % under 1 sun illumination, indicating its high potential as a photoelectrode material for the DSCs.

A Novel Approach for Detecting and Classifying Breast Cancer in Mammogram Images

Abstract: This study uses data mining techniques for computer-aided diagnosis that involves the feature extraction for cancer detection, so as to help doctors towards making optimal decisions quickly and accurately. Features play an important role in detecting the cancer in the digital mammogram and feature extraction stage is the most vital and difficult stage. In this paper, an enhanced feature extraction method named Multiscale Surrounding Region Dependence Method (MSRDM) is proposed to be effective in classifying the mammogram images into normal or benign or malignant. This proposed system is based on a four-step procedure: Regions of Interest specification, two dimensional discrete wavelet transformation, and multiscale surrounding region dependence matrix computation and feature extraction. The performance of the proposed feature set is compared with the conventional texture-analysis methods such as gray level cooccurence matrix features and surrounding region dependence method features. Experiments have been conducted on both real and benchmark data and the results have been proved to be progressive.

Parkinson's disease prediction using machine learning approaches

Abstract: This paper proposes the application of a Fully Complex-Valued Radial Basis Function network (FC-RBF), Meta-Cognitive Fully Complex-Valued Radial Basis Function network (Mc-FCRBF) and Extreme Learning Machine (ELM) for the prediction of Parkinson's disease. With the help of Unified Parkinson's Disease Rating Scale (UPDRS), the severity of the Parkinson's disease is predicted and for untreated patients, the UPDRS scale spans the range (0–176). The FC-RBF network uses a fully complex valued activation function sech, which maps cn → c. The performance of the complex RBF network depends on the number of neurons and initialization of network parameters. The implementation of the self-regulatory learning mechanism in the FC-RBF network results in Mc-FCRBF network. It has two components: a cognitive component and a meta-cognitive component. The meta-cognitive component decides how to learn, what to learn and when to learn based on the knowledge acquired by the FC-RBF network. Extreme learning mechanism uses sigmoid activation function and it works with fast speed. In ELM network, the real valued inputs and targets are applied to the network. The result indicates that the Mc-FCRBF network has good prediction accuracy than ELM and FC-RBF network.

Effective watermarking algorithm to protect Electronic Patient Record using image transform

Abstract: Due to development of latest technologies in the areas of communication and networking, the present businesses are moving to the digital world for effectiveness, convenience and security. Medical images require special safety and confidentiality because critical judgment is done on the information provided by medical images. Digital watermarking is an emerging technology to protect multimedia data for security purpose. This project suggests a comparative performance of digital image watermarking scheme using Discrete Cosine Transform (DCT) and Discrete Wavelet Transform (DWT) separately and their performance has been measured by using metrics like PSNR, Quality Index and Elapsed time. Initially, the Medical image is decomposed using image transforms like DCT or DWT. Subsequently, the watermark embedding and extraction process are to be performed in frequency domain transform along with LSB substitution algorithm which is of spatial domain. The performance of the proposed watermarking method is explained with the aid of experimental results.

Mechanical properties and water absorption of snake grass longitudinal fiber reinforced isophthalic polyester composites

Abstract: Natural fiber reinforced composites have replaced the existing conventional materials due to its light weight and enhanced load-bearing capabilities. In the present work, the newly identified snake grass (Sansevieria ehrenbergii) fiber-reinforced isophthalic polyester composites are prepared by simple hand lay-up method with different fiber weight fractions. The mechanical properties like tensile strength, flexural strength and modulus are analyzed for the longitudinal and transverse direction according to the prescribed standards. The obtained tensile strength and modulus are compared with the theoretically predicted values. The impact strength and energy absorption of the composites are analyzed and compared with control. The water uptake of pure and fiber incorporated resin under varying time period and climatic conditions are examined. The experimental results proves that the composites containing high fiber weight content contribute to remarkable increase in mechanical properties and water absorption...

Statistical modeling and optimization of alkaline protease production from a newly isolated alkalophilic Bacillus species BGS using response surface methodology and genetic algorithm.

Abstract: A new hyperactive alkalophilic bacterial strain (Bacillus sp. BGS) was isolated from samples collected from soil that received the effluent of a milk processing industry located in Madurai, Tamilnadu, India, and this bacterial strain was used for the production of alkaline protease. Four out of eight variables, such as molasses, peptone, pH, and inoculum size, have been identified through Plackett–Burman (PB) design and used for the alkaline protease production. These significant variables were further optimized through a hybrid system of response surface methodology (RSM) followed by genetic algorithm (GA). The optimal combination of media components and culture conditions for maximal protease production was found to be 16.827 g/L of peptone, 1.128% (v/v) of molasses, pH value of 11, and 2% (v/v) of inoculum size. A 6.36-fold increase in protease production was achieved through the RSM–GA hybrid system. The protease activity increased significantly with an optimized medium (2,992.75 U/mL) as opposed to a...

Machining fixture layout optimisation using genetic algorithm and artificial neural network

Abstract: The purpose of this research work is to design an optimum fixture layout in order to reduce the maximum elastic deformation of the workpiece caused by the clamping and machining forces acting on the workpiece while machining. First a genetic algorithm (GA) based optimisation procedure to solve the fixture layout optimisation problem is briefly explained, and then combined GA and artificial neural network (ANN) based optimisation procedure for fixture layout design is explained. In the combined GA and ANN approach, the resulting fixture layouts generated by GA are given as input to ANN and the maximum workpiece deformation for each fixture layout is found out by using ANN. The optimal fixture layout is the one which shows the minimum deformation among others. The results that are obtained by using GA and the combination of GA and ANN are compared.