Conference Papers
Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506
Browse
10 results
Search Results
Item Gender Detection using Handwritten Signatures(Institute of Electrical and Electronics Engineers Inc., 2018) Mohit Reddy, J.; Guru Pradeep Reddy, T.; Mishra, S.; Mulimani, M.; Koolagudi, S.G.In this paper, a method is proposed which uses both Image Processing and Machine Learning techniques which detects the gender of a person using handwritten signature. A photograph of a handwritten signature is given as input to the model which then extracts different features like pen pressure, slant angle, count external and internal contours etc. The features extracted from multiple images in the dataset are used to train the model, which then predicts the output of a new input given to it. Our objective is to collect unbiased datasets from a set of people and feed those signatures to the model, carrying out the statistical analysis and calculating the accuracy of the algorithm after every signature classification. We have used Adaboost classifier which gave a cross-validation accuracy of 73.2% compared to other classifiers like Gradient Boosting Classifier, Random Forest Trees and Multi-Layer Perceptron which gave 73.2%, 63.2% and 59.6% accuracies respectively. Copy Right © INDIACom-2018.Item Real and reactive power control of solar grid-tie inverter under distorted grid conditions(Institute of Electrical and Electronics Engineers Inc., 2018) Damodaran, R.; Ramana, V.V.; Balasubramanian, B.; Mishra, S.Due to increasing density of integration of distributed generation systems (DGS) to the utility grid, continuous power supply from DGS is being considered desirable during temporary grid faults, to provide support to the grid under conditions like voltage sags. Researchers have been focusing on improved topologies and control strategies in order to enhance the performance of solar grid-tied inverters, especially during grid faults. This paper presents a simple yet accurate control strategy for three-level four-wire grid-tie inverters for real and reactive power injection which operates satisfactorily under distorted grid conditions. Independent control of the input and output side is obtained by using two-stages, with boost converter obtaining maximum power from the solar panels, and inverter injecting this power into the grid with grid standards being met. A DC link balancing technique using proportional controller is also presented. A three-level neutral point clamped grid-tie inverter with the proposed control is modeled in MATLAB-Simulink platform and its effectiveness, in terms of accuracy of operation under non-ideal grid conditions and reduction in total harmonic distortion, is validated using the simulation results. © 2018 IEEE.Item Efficient global peak tracking of PV system under mismatching conditions using searching technique and bisection method(Institute of Electrical and Electronics Engineers Inc., 2018) Ramana, V.V.; Mudlapur, A.; Damodaran, R.; Balasubramanian, B.; Mishra, S.Rapidly increasing energy demand is leading to increase in installation of photovoltaic (PV) plants across the globe. The PV modules are subjected to various conditions like shading, soiling and degradation because of which the performance of PV panels deteriorates. This will affect the characteristics of PV module which may lead to several stairs in I-V curves and several peaks in P-V curve. In this paper the various causes of occurrences of mismatching conditions are analyzed. An efficient two stage algorithm that tracks global maximum power point (GMPP) accurately under such mismatching conditions is proposed in the paper. In the first stage, a simple searching technique and bisection method are used to bring the operating point to the proximity of global maximum. In the second stage, hill climbing method maintains the operating point at global peak. MATLAB Simulations and experimental verifications are performed for validating the proposed approach. © 2018 IEEE.Item Modified Hysteresis Current Control for Single Phase Solar Grid-tie Z-Source Inverter(Institute of Electrical and Electronics Engineers Inc., 2018) Damodaran, R.; Balasubramanian, B.; Mudlapur, A.; Ramana, V.V.; Mishra, S.This paper proposes a simple novel technique to determine the shoot-through states in a hysteresis controlled solar grid connected single phase Z-source inverter (ZSI). The ZSI along with the proposed controller exhibits high efficiency due to the single stage conversion. Fast and robust dynamic response can be obtained by using hysteresis control. Hence the proposed control for ZSI is ideal for grid integration of photovoltaic sources. The error current is used to determine shoot-through states based on calculation of current limits as a function of the boost ratio and modulation index. The maximum boost ratio, that can be obtained by the proposed control without causing low frequency ripples in Z-source network, is expressed as a function of modulation index. To verify the analysis, simulations are performed for input and grid voltage variations using MATLAB-Simulink platform. © 2018 IEEE.Item Single stage PV fed reduced inverter based PMSM for standalone water pumping application(Institute of Electrical and Electronics Engineers Inc., 2018) Karthikeyan, A.; Koothu Kesavan, K.K.; Varsha, S.; Venkatesa Perumal, B.; Mishra, S.This paper proposes a standalone PV fed PMSM drive for water pumping. The proposed system aims at reducing the switching losses and overall cost by using reduced switch inverter. The proposed system comprises of a PMSM drive, fed by PV source through an inverter employing reduced number of switches. The inverter uses only four switches whereas the conventional VSI utilizes six switches. Field oriented control scheme is employed to control the PMSM drive. Perturb and Observe (PO) MPPT technique is used to generate a speed reference to PMSM drive. Simulation studies for the proposed PV fed 400 W PMSM driven water pumping system are performed using MATLAB platform and respective results are presented to show the efficacy of the system. © 2018 IEEE.Item Output voltage control of DC-DC boost converter using model predictive control approach(Institute of Electrical and Electronics Engineers Inc., 2018) Khunte, K.V.; Madichetty, S.; Mishra, S.This article contains an approach to control the output voltage of dc-dc boost converter using model predictive control algorithm. State space modeling of a boost converter is done for both continuous and discontinuous conduction mode. By implementing online optimization and constraint satisfaction it produces a control input which regulates the output terminal voltage to its given reference. Disturbances produced due to load variation at output terminal are taken care of by an observer which is implemented by using kalman state estimator. Simulations are carried out on Simulink-MATLAB platform to examine the merits of proposed strategy, which includes fast transient response and robustness. © 2018 IEEE.Item A Transformerless Photovoltaic Microinverter using High-gain Z-Source Boost Converter for Single-phase Grid connected Applications(Institute of Electrical and Electronics Engineers Inc., 2020) Prabhu, N.S.; Damodaran, R.; Balasubramanian, B.; Mishra, S.Microinverters are the latest advancements in the area of photovoltaic (PV) technology due to their compact design. They improve energy efficiency of PV systems due to the reduced effect of non-uniform operating conditions caused primarily by shading. However due to the low output voltage rating of PV modules, high output to input voltage gain ratio is required for microinverters integrated to the utility grid. Several of the existing techniques use transformers or coupled inductors to achieve high gain ratios. These often require precise design and increase the size and cost of the PV system. The Z-source network based high-gain converters can offer superior performance with reduced components, size and cost. However their operation in a microinverter for grid connected applications have not yet been studied. Therefore this paper presents a novel microinverter utilizing a high-gain Z-source boost converter followed by an H-bridge inverter. The principle of operation of the microinverter is detailed and the overall gain ratio is obtained. The operating region in the I-V characteristics of PV module is determined and the control strategies implemented are discussed in detail. Simulations are performed in MATLAB/Simulink platform to validate the performance of the proposed microinverter under possible operating condition. © 2020 IEEE.Item SUPER-NATURALINSTRUCTIONS: Generalization via Declarative Instructions on 1600+ NLP Tasks(Association for Computational Linguistics (ACL), 2022) Wang, Y.; Mishra, S.; Alipoormolabashi, P.; Kordi, Y.; Mirzaei, A.; Arunkumar, A.; Ashok, A.; Dhanasekaran, A.S.; Naik, A.; Stap, D.; Pathak, E.; Karamanolakis, G.; Lai, H.G.; Purohit, I.; Mondal, I.; Anderson, J.; Kuznia, K.; Doshi, K.; Patel, M.; Pal, K.K.; Moradshahi, M.; Parmar, M.; Purohit, M.; Varshney, N.; Kaza, P.R.; Verma, P.; Puri, R.S.; Karia, R.; Sampat, S.K.; Doshi, S.; Mishra, S.; Reddy, S.; Patro, S.; Dixit, T.; Shen, X.; Baral, C.; Choi, Y.; Smith, N.A.; Hajishirzi, H.; Khashabi, D.How well can NLP models generalize to a variety of unseen tasks when provided with task instructions? To address this question, we first introduce SUPER-NATURALINSTRUCTIONS, a benchmark of 1, 616 diverse NLP tasks and their expert-written instructions. Our collection covers 76 distinct task types, including but not limited to classification, extraction, infilling, sequence tagging, text rewriting, and text composition. This large and diverse collection of tasks enables rigorous benchmarking of cross-task generalization under instructions-training models to follow instructions on a subset of tasks and evaluating them on the remaining unseen ones. Furthermore, we build Tk-INSTRUCT, a transformer model trained to follow a variety of in-context instructions (plain language task definitions or k-shot examples). Our experiments show that Tk-INSTRUCT outperforms existing instruction-following models such as InstructGPT by over 9% on our benchmark despite being an order of magnitude smaller. We further analyze generalization as a function of various scaling parameters, such as the number of observed tasks, the number of instances per task, and model sizes. We hope our dataset and model facilitate future progress towards more general-purpose NLP models. © 2022 Association for Computational Linguistics.Item High-gain DC-DC converter with zero input ripple current : Design and Analysis*(Institute of Electrical and Electronics Engineers Inc., 2023) Mishra, S.; Shetty, S.; Vinatha Urundady, U.In this paper, a non-isolated high-gain dc-dc converter that utilizes switched-capacitor and switched-inductor (SC-SL) network is proposed and thoroughly analyzed. The proposed topology features a single switch and less number of passive elements as compared to recently emerged high-gain converters. The mathematical analysis of the proposed converter is carried out to find the converter voltage gain and stresses on power devices.The converter achieves a gain of nine times at 50% duty cycle with comparatively less voltage stress on power devices. Additionally, the converter encompasses the current mirror ripple cancellation circuit (CMRCC) to eliminate input current ripples. The converter is modelled and verified in continuous conduction mode(CCM) using MATLAB/SIMULINK. The obtained findings exhibit that the input current ripples are effectively eliminated by the CMRCC implementation. © 2023 IEEE.Item Modelling of PEM fuel cells : A Parameterized Approach(Institute of Electrical and Electronics Engineers Inc., 2023) Shetty, S.; Bharath, Y.K.; Mishra, S.; Vinatha Urundady, U.PEM fuel cells are encouraging energy conversion technologies that offer low emissions and high efficiency for various applications, auto motives in particular. In order to create flexible and adaptive models that accurately depict the complex electrochemical, transport, and thermodynamic processes within these systems, a parameterized technique for modelling PEM fuel cells is essential. This work focuses on investigating the profound impact of load and temperature variations on the steady-state behavior of PEM fuel cells. The parameterized approach is then employed to develop a dynamic model of PEM fuel cells. The dynamic model captures the intricate dynamics of the cell under varying load conditions, considering the transient behavior of the cell components and the interactions between the different physical phenomena. © 2023 IEEE.