Nishchal K. Verma

Robust Wide-Area TS Fuzzy Output Feedback Controller for Enhancement of Stability in Multimachine Power System

This paper presents a systematic procedure for designing a wide-area centralized Takagi-Sugeno fuzzy controller to improve the angular stability of a multimachine power system. The proposed fuzzy controller is robust and designed by satisfying certain linear matrix inequality conditions, to stabilize the system at multiple operating points. The bilinear matrix inequality problem, encountered in Lyapunov-based stability criterion, has been converted into a convex optimization problem to eliminate iterative solution. The input-output control signals are selected in the proposed control scheme by defining joint model controllability and observability index applying geometric approach. The proposed wide-area control scheme employs a global signal from the centralized controller to damp out the interarea mode of oscillations, apart from the local controllers, which are assumed to be present to damp out the local mode of oscillations. The proposed control scheme has been implemented on three test systems. The effectiveness of the proposed control scheme has been compared with a robust wide area control scheme based on mixed H2/H∞ output feedback synthesis and with a conventional power system stabilizer control scheme.

Content Based Color Image Classification using SVM

We propose a novel approach for content based color image classification using Support Vector Machine (SVM). Traditional classification approaches deal poorly on content based image classification tasks being one of the reasons of high dimensionality of the feature space. In this paper, color image classification is done on features extracted from histograms of color components. The benefit of using color image histograms are better efficiency, and insensitivity to small changes in camera view-point i.e. translation and rotation. As a case study for validation purpose, experimental trials were done on a database of about 500 images divided into four different classes has been reported and compared on histogram features for RGB, CMYK, Lab, YUV, YCBCR, HSV, HVC and YIQ color spaces. Results based on the proposed approach are found encouraging in terms of color image classification accuracy.

SVM based methods for arrhythmia classification in ECG

In this study, Support Vector Machine (SVM) based methods have been used to classify the electrocardiogram (ECG) arrhythmias. Among various existing SVM methods, three well-known and widely used algorithms one-against-one, one-against-all, and fuzzy decision function are used here to distinguish between the presence and absence of cardiac arrhythmia and classifying them into one of the arrhythmia groups. The various types of arrhythmias in the Cardiac Arrhythmias ECG database chosen from University of California at Irvine (UCI) to train SVM, include ischémie changes (coronary artery disease), old inferior myocardial infarction, sinus bradycardy, right bundle branch block, and others. The results obtained through implementation of all three methods are thus compared as per their accuracy rate in percentages and the performance of the SVM classifier using one-against-all (OAA) method was found to be better than other techniques. ECG arrhythmia data sets are of generally complex nature and SVM based one-against-all method is found to be of vital importance for classification based diagnosing diseases pertaining to abnormal heart beats.

A Coherency-Based Approach for Signal Selection for Wide Area Stabilizing Control in Power Systems

In this paper, a systematic procedure to select wide-area input/output control signals based on a signal coherency approach to damp out inter-area mode of oscillations is proposed. The coherent signal groups are chosen based on a data clustering approach. The input/output signal selection is carried out in two steps. First, the data is transformed into orthogonal space to make correlated variables uncorrelated, by applying principal component analysis. Then, the principal component vectors are given as input to a self-organizing map for final data clustering. From each clustered signal group, a signal out of those with common features is selected. For clustering, the data are collected from the system dynamic simulation considering few critical line contingency. The efficacy of the proposed feature selection based method is verified by comparing the results with a geometric-based approach and a pole vector direction based signal selection approach on 39-bus New England and 68-bus New England New York power systems.

Finding sensitive sensor positions under faulty condition of reciprocating air compressors

A computer based data acquisition (DAQ) system plays an important role in Health Monitoring of any machine. Health Monitoring of a reciprocating air compressor using a computer based DAQ system and timely identification of potential faults can prevent failures. In this paper we have employed a graphical user interface for data acquisition that allows flexible acquisition and processing of analog and digital data. The first stage of Health Monitoring procedure is dedicated to collection of all necessary information about the machine. Our research is focused for finding sensitive positions. Also we have compared sensitive positions under healthy condition with the faulty condition.

Additive and Nonadditive Fuzzy Hidden Markov Models

We present a novel approach for the development of fuzzy hidden Markov models (FHMMs) by exploiting both additive and nonadditive properties of input fuzzy sets in the fuzzy rules of generalized fuzzy model (GFM). This development utilizes 1) Gaussian mixture model (GMM) to manipulate the mixture parameters for the input fuzzy sets and 2) GFM rules for the inclusion of states in the consequent part to be able to use HMM. Taking the components of Gaussian mixture density conditioned on the past system states and making use of equivalence of GMM with GFM, parameters of the additive and nonadditive FHMMs are estimated using the forward-backward procedure of the Baum-Welch algorithm. The additive and nonadditive FHMMs are validated on three benchmark applications involving time-series prediction, and the results are compared and found to be better than or equal to those of the existing recent fuzzy models.

Intelligent Condition Based Monitoring Using Acoustic Signals for Air Compressors

Intelligent fault diagnosis of machines for early recognition of faults saves industry from heavy losses occurring due to machine breakdowns. This paper proposes a process with a generic data mining model that can be used for developing acoustic signal-based fault diagnosis systems for reciprocating air compressors. The process includes details of data acquisition, sensitive position analysis for deciding suitable sensor locations, signal pre-processing, feature extraction, feature selection, and a classification approach. This process was validated by developing a real time fault diagnosis system on a reciprocating type air compressor having 8 designated states, including one healthy state, and 7 faulty states. The system was able to accurately detect all the faults by analyzing acoustic recordings taken from just a single position. Additionally, thorough analysis has been presented where performance of the system is compared while varying feature selection techniques, the number of selected features, and multiclass decomposition algorithms meant for binary classifiers.

Intelligent condition based monitoring of rotating machines using sparse auto-encoders

Support Vector Machine (SVM) has been very popular for use in machine fault diagnosis as classifier. In most of the complex machine learning problems, the main challenge lies in finding good features. Sparse autoencoders have the ability to learn good features from the input data in an unsuperivised fashion. Sparse auto-encoders and other deep architectures are already showing very good results in text classification, speaker and speech recognition and face recognition as well. In this paper, we compare the performance of sparse autoencoders with soft max regression, fast classifier based on Mahalanobis distance and SVM in fault diagnosis of air compressors.

Two-tiered wireless sensor networks - base station optimal positioning case study

In this study, the question of `Where should the base station be placed in a two-tiered wireless sensor network (WSN) field?` has been investigated. The objective is to minimise the overall energy consumption in a WSN. A heuristic algorithm has been proposed to find such a base station location. Considering some nodes to be far enough to use a different path loss model for their signals to the base station, our proposed algorithm considers two categories of nodes and hence two different path loss models based on their distance from the base station. The results show that the new algorithm provides a better base station location than the earlier methods. The overall energy consumption is quite close to the optimal solution.

Object Matching Using Speeded Up Robust Features

Autonomous object counting system can help industries to keep track of their inventory in real time and adjust their production rate suitably. In this paper we have proposed a robust algorithm which is capable of detecting all the instances of a particular object in a scene image and report their count. The algorithm starts by intelligently selecting Speeded Up Robust Feature (SURF) points on the basis stability and proximity in the prototype image, i.e. the image of the object to be counted. SURF points on the scene image are detected and matched to the ones on the prototype image. The notion of Feature Grid Vector (FGV) and Feature Grid Cluster (FGC) is introduced to group SURF points lying on a particular instance of the prototype. A learning model based on Support Vector Machine has been developed to separate out the true instances of the prototype from the false alarms. Both the training and inference occur almost in real time for all practical purposes. The algorithm is robust to illumination variations in the scene image and is capable of detecting instances of the prototype having different distance and orientation w.r.t. the camera. The complete algorithm has been embodied into a desktop application, which uses a camera feed to report the real time count of the prototype in the scene image.