Kalyan Chatterjee

Mixed Approach of Order Reduction for Single-Input Single-Output (SISO) Systems

In this paper, a mixed method is proposed which combines the improved Pade approximations and the eigen spectrum analysis for reducing the higher-order system. In this method, system stiffness and pole centroid of both original and reduced-order system remain same. The denominator of higher-order system (HOS) is derived using the eigen spectrum analysis, and the numerator is derived by using improved Pade approximation. The latter method of order reduction utilizes both time moments and Markov parameters. The stability and quality of the reduced-order system are compared with the existing methods of order reduction. To understand the proposed method, paper includes some numerical examples of single-input single-output (SISO) systems.

SISO Method Using Modified Pole Clustering and Simulated Annealing Algorithm

A mixed method by using the modified pole clustering technique and simulated annealing is proposed to reduce higher-order mathematical model into a smaller one. The denominator and numerator polynomials are obtained by using a modified pole clustering technique and simulated annealing algorithm, respectively. The proposed-biased method generates k number of reduced models from higher-order systems. The compatibility of the method has been checked via time responses of the original higher-order system and the reduced-order system, respectively. Also, the proposed method has been compared with few known model-order reduction techniques through performance indices.

A Novel Adaptive Feature Extraction for Detection of Cardiac Arrhythmias Using Hybrid Technique MRDWT & MPNN Classifier from ECG Big Data

Abstract The efficient automatic detection of cardiac arrhythmia using a hybrid technique from ECG big data has been proposed with novel feature extraction technique using Multiresolution Discrete Wavelet Transform (MRDWT) and Multilayer Probabilistic Neural Network (MPNN) classifier. Big Data of ECG signals have been selected from MIT–BIH arrhythmia database for detection of two types of arrhythmias LBBB (Left Bundle Branch Block) and RBBB (Right Bundle Branch Block). The proposed technique can accurately detect and classify LBBB and RBBB along with normal heartbeat. A novel and hybrid method of detection of cardiac arrhythmia have four main stages: denoising of raw ECG, baseline wander removal, proposed feature extraction, and detection of abnormal heartbeats using MPNN neural classifier. 8600 ECG beats were selected, including 4200 normal and 4400 abnormal beats (2200 LBBB and 2200 RBBB) were utilized for testing the proposed technique. The detection outcome using MPNN was compared with other two neural classifiers: Feed Forward Neural Network (FFNN) and Back Propagation Neural Network (BPNN) classifiers. The accuracy and efficiency of classifiers performance were attained in terms of CER (Classification Error Rate), SP (Specificity), Se (Sensitivity), Pr (Precision), PPr (Positive Predictivity) and F-Score. The system performance is achieved with 96.22%, 97.15% and 99.07% overall accuracy using FFNN, BPNN and MPNN. The average percentage of classification error rate (CER) using MPNN classifier is lowest 0.62% whereas FFNN and BPNN show 2.2% and 1. 90% average CER.

A unique feature extraction using MRDWT for automatic classification of abnormal heartbeat from ECG big data with Multilayered Probabilistic Neural Network classifier

Abstract This paper employs a novel adaptive feature extraction techniques of electrocardiogram (ECG) signal for detection of cardiac arrhythmias using multiresolution discrete wavelet transform from ECG big data. In this paper, five types ECG arrhythmias including normal beats have been classified. The MIT-BIH database of 48 patient records is utilized for detection and analysis of cardiac arrhythmias. Proposed feature extraction utilizes Daubechies as wavelet function and extracts 21 feature points which include the QRS complex of ECG signal. The Multilayered Probabilistic Neural Network (MPNN) classifier is proposed as the best-suited classifier for the proposed feature. Total 1700 ECG betas were tested using MPNN classifier and compared with other three classifiers Back Propagation (BPNN), Multilayered Perceptron (MLP) and Support Vector Machine (SVM). The system efficiency and performance have been evaluated using seven types of evaluation criteria: precision (PR), F-Score, positive predictivity (PP), sensitivity (SE), classification error rate (CER) and specificity (SP). The overall system accuracy, using MPNN technique utilizing the proposed feature, obtained is 99.53% whereas using BPNN, MLP and SVM provide 97.94%, 98.53%, and 99%. The processing time using MPNN classifier is only 3 s which show that the proposed techniques not only very accurate and efficient but also very quick.

Assessment of MPPT Techniques During the Faulty Conditions of PV System

The contribution of Distributed Generation (DG) systems like wind energy systems and solar Photovoltaic (PV) systems on the generation of electricity has increased. Out of these DG systems, the PV systems have gained wide popularity, because of the availability of solar energy throughout the day. Depending on the size of PV installations, a large number of PV modules can be interconnected in the form of series and parallel connection. Since a large number of modules are interconnected, it is possible for the faults in a PV array to occur due to the failure of protection system, which can cause damage to the PV module and also the decrease in the output power. This paper presents the tracking of a maximum power point under the faulty conditions of 12x5 PV array. The fault conditions that have been considered in the PV array are open circuit fault, line to ground, line to line and failure of bypass diodes. Perturb and observe, incremental conductance and fuzzy logic controller are the maximum power point tracking techniques that have been implemented. For each of the fault conditions, the results have been presented in terms of the maximum power tracked, tracking time and tracking efficiency.

Reliability analysis of a multi-state wind farm using Markov process

Abstract The usage of wind power is much more popular nowadays although the intermittent nature of wind makes the wind power generation unreliable and stochastic to a large extent. With the uncertainty of wind, the power system planners are faced with a big challenge to design a sound model of power systems. So, reliability estimation of a wind farm has become a great challenge for the researcher. In this article, Markov Reward method has been proposed for developing a realistic model of a wind farm in respect to the wind farm relative availability function. According to the proposed model, the performance of a wind farm has been assessed in terms of different reliability indices. The novelty of this model is that it can be capable of counting any level of power generation made by the wind farm irrespective of failure and success condition reported in the previous literature. A comparison of this study reveals the efficacy of the proposed model over the existing General Markov Reward method. From a case study, the practical validation of this model has been carried out.

PEMFC connected in standalone mode with five level inverter

To produce useful form of energy without causing any harm to the environment is a continuing challenge. This puts human-kind in a pursuit of green energy sources, one among which is the fuel cell technology. Fuel cells has always been an element of interest for the past few decades. But the technology is still in its crib. Research is still being carried out to avail it for daily use at a cheaper cost. In this paper, prior interests lie in utilizing fuel cells by connecting them to an individual three-phase load and study its performance.