Power System Stability Analysis of Single Machine Infinite Bus System using Firefly Algorithm
Abstract: With the growth of interconnected power systems and particularly the deregulation of the industry, difficulties related to low frequency oscillation have been widely reported, together with major incidents. As the most economical damping controller, power system stabilizer (PSS) has been widely used to suppress the low frequency oscillation and enhance the system dynamic stability. Traditional methods for determining PSS placements are based on the analysis of the interconnected system. Though, the design of the PSS is based on a simplified single machine infinite bus (SMIB) model. Traditional methods for determining PSS placements are based on the analysis of the interconnected system. In this paper, the design of the PSS is based on a simplified single machine infinite bus (SMIB) model using Firefly Algorithm. SMIB with PID controller is also implemented.

Authors: Shreya Namdev Nidhi Khurpiya

File Name: Shreya_50800-17-211.pdf
 Download Abstract | Download Full Paper

Classification of Heart Disease using Neural Network Classifier
Abstract: Heart disease prediction is treated as most complicated task in the field of medical sciences. Thus there arises a need to develop a decision support system for detecting heart disease of a patient. Data mining techniques have been widely used in clinical decision support systems for prediction and diagnosis of various diseases with good accuracy. The main objective of this research work is to develop a prototype which can determine and extract unknown knowledge (patterns and relations) related with heart disease from a past heart disease database record. This paper proposes back propagation Neural Network technique for heart disease prediction. Performance of proposed approach is evaluated using confusion matrix plot.

Authors: Shiva Shrivastava, Neeraj Mehta

File Name: Shiva_50800-17-210.pdf
 Download Abstract | Download Full Paper

Facial Expression Recognition using Hybrid Method of Local Binary Pattern and Gabor Filter Features with Random Forest Classifier
Abstract: The face though seems an easy object to be recognized by retina but the artificial intelligence is not yet intelligent enough to do the task easily. As the source of a face is generally an image capturing object, there are lot of variations and complexions that persists with the image like (for example: noise, rotation etc.). There are many techniques that use some or other algorithm to find similarity in face model and the test image and most of them are successful on their part to attain better test similarities. However, considering the diverse scale of applications and mode of image sourcing, a single algorithm cannot get maximum efficiency everywhere. Even after using the best algorithm for a particular task, an application has to counter with challenges of face recognition. The main aim of this paper is to analyze the Hybrid method of Local Binary Pattern and Gabor Filter features and its performance when applied to facial expression recognition. This algorithm creates a subspace (face space) where the faces in a database are represented using a reduced number of features called feature vectors and Random Forest Classifier calculates the similarity score for performance evaluation which will provide improved results in terms of recognition accuracy.

Authors: Priyanjali Kuruvila, Rasna Sharma

File Name: Priyanjali_50900-17-102.pdf
 Download Abstract | Download Full Paper

Study of Shunt Active Power Filter for Harmonic Reduction Using Different Methods
Abstract: The nonlinear loads generate harmonic currents which distorts the power quality in the distribution network, so it can be considered as a pollutant which pollutes the entire power system. To overcome problems due to harmonics, Shunt Active Power Filters are used as a solution. This paper represents various methods to eliminate current harmonics caused by the non-linear loads and compensate reactive power has been presented in this work. Modularity and simple maintenance make the proposed SAPF an attractive solution compared to some conventional configurations.

Authors: Mohd Junaid Mansoori, Prakash Bahrani, Nitesh Agrawal, Nasir Hussain Mansoori

File Name: Mansoori_50900-17-105.pdf
 Download Abstract | Download Full Paper

ECG Signal Classification using PCA, DWT and Neural Network Classifier
Abstract: This paper offers ECG signal classification system using Principal Component Analysis (PCA) technique to reduce the dimensionality of test signal. Discrete Wavelet Transform (DWT) is used for feature extraction. Power Spectral Density (PSD) is another feature for the spectrum of ECG. This process helps in enhancing the classification accuracy. Classification is done using Neural Network classifier. In this paper, the signal processing and neural network toolbox are used in MATLAB environment. The processed signal source came from the Massachusetts Institute of Technology Beth Israel Hospital (MIT-BIH) arrhythmia database which was developed for research in cardiac electrophysiology.

Authors: Ankish Dangi, Dr. G. D. Gidwani

File Name: Ankish_50900-17-107.pdf
 Download Abstract | Download Full Paper