http://localhost:8081/jspui/handle/123456789/28 2026-05-07T21:26:44Z http://localhost:8081/jspui/handle/123456789/20486 Title: HEALTH MONITORING AND IDENTIFICATION OF INDUCTION MOTOR FAULTS BASED ON CURRENT AND VIBRATION SIGNALS Authors: Kumar, Rajeev Abstract: Induction motors, especially three-phase induction motors, play an important role in various industries owing to their advantages over other electrical machines due to their reliability and safe operation. However, if any faults or failures occur in the motor, it can lead to increased breakdown time and substantially generate huge losses in terms of revenue and maintenance. Therefore, an early fault detection scheme is needed for prior maintenance of these motors. In the current scenario, the demand for health monitoring of induction motors is growing with the objective of reducing their operational costs, enhancing their operational reliability and providing better service to customers. In recent trends, various monitoring methods are being used to investigate fault conditions in induction motors. On the basis of these methods, the health monitoring schemes of induction motors are categorised into two major areas such as feature extraction based analysis and knowledge based analysis. For real time health monitoring, signal processing based feature extraction methods are being used. The incipient faults are diagnosed by extracting signatures from measured signals such as vibration signal, noise signal, current and voltage signal etc. The knowledge based approaches are quite popular among researchers, which are being developed by using artificial intelligence, deep learning and machine learning algorithms to detect the faults most accurately at early stages. The research presented in this thesis is primarily focused on signal processing analysis, using machine learning algorithms to identify stator winding inter-turn faults and ball bearing faults. To identify these faults, experimental setups are developed for each fault separately that are capable of measuring both current and vibration signals by conducting experiments in the laboratory. The subsequent analysis is carried out using MATLAB and Python programming. These early stage incipient faults are analysed utilising through current and vibration signals with various signal processing techniques such as Fast Fourier Transform (FFT), Park’s Vector Magnitude Analysis (PVM), Signal Envelope Identification Analysis (SEI) and Zero Crossing Time Detection (ZCT). 2024-05-01T00:00:00Z http://localhost:8081/jspui/handle/123456789/20478 Title: STATE ESTIMATION AND UNCERTAIN POWER FLOW OF ISLANDED AC AND HYBRID AC/DC MICROGRIDS Authors: Surendra, Kundukuri Abstract: The Distribution System (DS) has gained significant attention in power system studies due to the integration of innovative technologies such as distributed generations from renewable energy sources (RES), internet and communication technologies (ICT), advanced monitoring technologies (AMI, smart meters), and customer participation for demand side management. The power system studies can be broadly categorized into operational and planning studies. State Estimation (SE) and Power Flow (PF) are the most critical components in operational and planning studies. These two functions (SE and PF) are crucial as other operational and planning functions heavily rely on the results of SE and PF, such as fault detection, load balancing, and system stability analysis. The generation from renewable sources (RES) is increasing rapidly to meet the global goals of carbon neutrality, sustainable energy for the future, and eco-friendly mobility solutions. The RES integrated into DS is considered distributed generations (DGs), and they increase the efficiency of DS through reduced transmission losses. Further, the smaller size and less commissioning time reduce the capital cost of DS. The DGs increase the DS sustainability and reliability through Microgrid (MG) structures. MG is a portion of DS with a group of DGs, loads, and storage systems operating as a single control entity. Microgrids (MGs) are classified into AC, DC, and Hybrid AC/DC based on their distribution nature. These AC, DC, and hybrid AC/DC MGs can operate in two distinct modes: grid-connected and islanded. In the former, with the assumption of an infinite bus, the DGs in the DS aim to maximize the utilization of RES, and the intermittency of renewable DGs can be managed by the primary grid. However, in the face of natural calamities, grid-side faults, power quality issues, or cyber-attacks, the AC, DC, and hybrid AC/DC MGs switch to islanded mode. In this scenario, the primary objective is to find the equilibrium operating point to supply the critical loads through RES or conventional non-renewable generations. This independent operation of MGs not only enhances the reliability of the DS but also ensures the supply of power to customers even in the absence of the grid. Due to the absence of a slack bus for islanded AC, DC, and hybrid AC/DC MGs, the generation output from the DGs depends on the load requirements, with the load being shared proportionally to their maximum capacity. This approach is simple and cost-effective because it does not require any communication systems, which are prone to single-point failures. The amount of power generated by each DG in islanded AC MG (IMG) and islanded AC/DC hybrid MG (IHMG) is calculated using the droop characteristics of DGs. This thesis focuses on developing State Estimation (SE) and Uncertain Power Flow (UPF) models for Islanded AC Microgrids (IMG) and Islanded Hybrid AC/DC Microgrids (IHMG). 2024-06-01T00:00:00Z http://localhost:8081/jspui/handle/123456789/20477 Title: ANALYSIS AND DEVELOPMENT OF SINGLE STAGE SOLAR PV INVERTER SYSTEM Authors: Jaga, Om Prakash Abstract: T HEgrowingelectricitydemandisrisingduetotherisingpopulationanddepletionof non-renewableenergysources,rapidclimatechangeandglobalwarmingareamongthe major threats to today’sworld. Renewable energy sources (RESs) such as solar photovoltaic(PV)havebecomeanalternativeoptionforrisingenergydemands.Solarenergy iscountedamongsteasily availableRESs.One ofthemajoradvantages ofsolarPVis thatitis independentofend-userlocation.Inthecaseofnon-electrifiedremote areas,itbecomesmore economical to use solar photovoltaic (PV) based power generation compared to the conventional grid-connectedsystem.Critical loadapplications suchashealthcarefacilities, informationtechnology-enabledservices, telecommunicationnetworkingandelectronicloads likecomputersrequireanuninterruptedpowersupply.ThePVelectricityisnotaccessibleat night or during rainyweather.Grid-tied solar energyconversion system(SECS) delivers electricitytoessentialloads.However,gridoutagesarerelativelyfrequentinIndia,loweringthe dependabilityofgrid-tiedSECS.Thebatteryenergystorage(BES)systemcanbeinterfaced with/withoutabidirectionalDC/DCconverterattheDClinktoimprovegrid-tiedSEC system reliability.TheSECsystemcanbeclassifiedintotwomaincategories: two-stageandsingle stage. GridoutagesisfrequentandverycommoninIndia.Therefore, theutilitymainsmustbe isolatedseamlesslysoasnot toaffect critical loads and theoccurrenceof voltage/current sag/swellshouldnotaffectsystemperformance.Unintentionalgrid-outagemayoccurdueto theutilitymainsfailure,circuitbreakeropening,orhumanmistake.Henceforth,anislanding detection(ID)/re-synchronizationcontroller is requiredtobemadeoperational. Ingrid-tied mode, theACmainsmaintain the amplitude and frequencyof gridvoltage.However, in islandedmode, it isessential tomaintainloadvoltageamplitude, frequencyandTHDwithin thelimits inlinewithIEEE-519.Non-linearloadsconnectedatthepointofcommoncoupling pollute theutilitymainsby injectingharmonics.Harmonics in thegridcurrentmaycause overheating, short lifespans, andmalfunctioningof appliances.Thepower qualityofgrid currentistobesustainedwithinalimitinaccordancewithIEEE-519.Therefore,acontrolled single-phaseVSCalsooperatesasaDSTATCOMtocatertotheload'sreactivepowerdemand. Inthisthesiswork,dual-modetwo-stageandsingle-stagemultifunctional self-reliantgrid interactive single-phase SEC-BES systems are designed, controlled, simulated and implemented.Theprimegoalof thisworkis tosupplycontinuouspower tothecritical load i regardlessof theavailabilityof theACmains. Insurpluspowermode(SPM), thegenerated power fromthePVishigher thanthedemandpower fromthecritical loads.Therefore, the surpluspowerofthePVisusedtochargetheBESsystemandtheremainingactivepoweris injectedintotheACmains. Indeficitpowermode(DPM), thegeneratedpowerof thePVis lesserthanthedemandpowerofthe criticalloads.Insuchacase,thebalancepowerissupplied either fromtheACmainsor theBESsystem,whichdependsontheavailabilityof theAC mains. Thisworkmainlyclassifies the grid-tiedSEC-BES systemcontrol scheme into three categories.Inthefirstcategory,aPI-Leadcompensator-baseddual-loopandmodelpredictive controlschemesfor abidirectional DC/DCconverteraredesigned,simulatedandimplemented tocontrol theDCbusvoltageandBESsystemcharging/dischargingcurrentundervarious operatingmodes.Inaddition,thepresentedDC/DCconvertercontrolschemealsoensuresthe extractofmaximumpowerfromthePVarray.Moreover,astate-of-chargecontrollerwitha bidirectionalDC/DCconverterprotectstheBESsystemfromdeep-discharging/overcharging. ThestabilityanalysisofabidirectionalDC/DCconverterhasbeenpresentedthoroughlyboth for the PI-lead compensator and model predictive controller. Designing the PI-Lead compensator parametersbasedon systemstabilityconstrained tooperate thebidirectional DC/DCconverter inastableregionundervariousoperatingconditionsbecomes important. Therefore, thedual-loopPI-Leadcontroller parametersaredesignedbasedon the stability constraintsinthefrequency domain. Thestabilityanalysis ofabidirectionalDC/DCconverter for amodel predictive controller is analysed in the Z-domain. The Jacobianmatrix is constructedusingthestatevariableequationsofabidirectionalDC/DCconverter.TheJacobian matrixeigenvaluesdeterminethestabilityofabidirectionalDC/DCconverter. Ifeigenvalues liesinsidetheunitecirclethentheDC/DC converteroperates inastableregion.Otherwise,the systemisunstable. ThebidirectionalDC/DCconverterparameters (DCbus capacitor and converter inductance)andtheweightingfactorforamodelpredictivecontrolleraredesigned basedonbidirectional DC/DCconverteroperation ina stability region. 2024-06-01T00:00:00Z http://localhost:8081/jspui/handle/123456789/20476 Title: ECG-BASED HUMAN IDENTIFICATION AND AUTHENTICATION Authors: Begum, Ritu Nazneen Ara Abstract: Identity theft is a major concern today. It has serious ramifications beyond data and personal information loss, necessitating the implementation of robust and efficient user identification systems. Identification and authentication of users is foremost for granting access to personal devices, accounts or confidential files. Traditional techniques include PINs, passwords, and tokens, which are viable to be forgotten, stolen or cracked. A fundamental development in the field of identity theft protection is biometrics, which is a reliable alternative to classical identification and authentication approaches. Of all the biometrics, the Electrocardiogram (ECG) has been on a roll due to its inevitably desirable intrinsic features that are not available in other biometric traits. Unlike other biometrics, it cannot be copied (signature), imitated (voice), duplicated (finger prints, iris etc.) and faked (gait). Moreover, it is the proof of aliveness of the person. ECG biometrics identification has been implemented using time domain, frequency domain, hybrid of the two, and very recently, the Deep Learning (DL) approaches. Recent state-of-the-art techniques involve CNNs for ECG based user identification. However, the CNN based methodologies have limitations of long training time and large datasets. The current research work is divided into three stages. First, a dataset of diverse ECG waveforms was created. The dataset was carefully built to include recordings from both genders, a wide age range, various signal-acquiring setups, different health and physical conditions, various physical postures and mental conditions. The signals in the dataset were prepared to be used with deep learning algorithms. This involved normalizing the amplitude and period, filtering, segmenting, and ultimately converting them into images. Second, several deep learning algorithms were explored and experimented using the prepared dataset, and the results were analysed. Two novel DL models have also been developed that address the objectives of increased identification rate (IDR) and decreased computational effort. Third, a cancellable ECG-biometric has been designed for secured ECG-based human authentication. The main aim of this study is to create an automated biometric system for user identification and authentication using ECG waveforms. In this investigation, an initial examination encompassed four distinct deep learning (DL) algorithms for the purpose of user identification utilizing the provided dataset. These DL algorithms comprise an LSTM network, a transfer learning approach employing ResNet-50, a customized residual network, and a customized dense network. Subsequent analysis of the findings revealed the dense network to be the most suitable for the given task. Extensive investigations were conducted on the Dense network for four distinct architectures, while meticulously analysing the activations of each layer. Furthermore, experiments were undertaken to assess the impact of reduced training dataset and multiple session ECG recordings on the network's performance. The current study also developed two deep learning (DL) models for user identification based on ECG data. The first model is an ensemble of UNet with ANN. The UNet part of the model is specifically designed for segmenting and extracting crucial features from the ECG waveform, while the ANN is iii iv responsible for classification. This model has also been implemented using TensorFlow/Keras and is suitable for deployment on portable devices with limited capacity. The proposed ensemble model generates a considerable number of learnable parameters, leading to escalated computation costs and enhanced hardware requirements. Consequently, a hybrid model comprising a dense CNN with a fire module was formulated. The implementation of the fire module within the network resulted in reduction of the learnable parameters by a considerable factor, thereby decreasing the computational effort, memory demand, and training duration. However, this reduction in learnable parameters has an impact on the model's performance. A study has also been conducted to develop a human authentication system that takes into account hardware limitations and computation costs. A simple yet effective ECG secret key generation technique for user authentication has been developed based on multiple recursive algorithm (MRA). The proposed keys are generated at two stages, with one part stored in the device and the other part generated on-the-fly. Concatenating both parts generates a 128-bit secret key that can be used for authentication. The keys generated with this technique are evaluated using metrics such as reliability, robustness, and entropy. The analysis of the results reveals that the proposed key generation technique can provide sufficient strength and randomness to the keys, ensuring a secure ECG-based user authentication system. 2024-06-01T00:00:00Z