Signal Processing and Soft Computing Approaches to Power Signal Frequency and Harmonics Estimation (Record no. 74377)

000 -LEADER
fixed length control field nam a22 7a 4500
005 - DATE AND TIME OF LATEST TRANSACTION
control field 20170712151927.0
006 - FIXED-LENGTH DATA ELEMENTS--ADDITIONAL MATERIAL CHARACTERISTICS--GENERAL INFORMATION
fixed length control field t||||||m 00| 0
008 - FIXED-LENGTH DATA ELEMENTS--GENERAL INFORMATION
fixed length control field 170712b2011 xxu||||| m||| 00| 0 eng d
041 ## - LANGUAGE CODE
Language code of text/sound track or separate title eng
100 ## - MAIN ENTRY--AUTHOR NAME
Personal name Ray , Pravat Kumar
Roll Number/Miscellaneous information 507EE004
245 ## - TITLE STATEMENT
Title Signal Processing and Soft Computing Approaches to Power Signal Frequency and Harmonics Estimation
260 ## - PUBLICATION, DISTRIBUTION, ETC. (IMPRINT)
Year of publication 2011
300 ## - PHYSICAL DESCRIPTION
Number of Pages 261 p.
502 ## - DISSERTATION NOTE
Degree Type Thesis (Ph.D)-
Name of granting institution National Institute of Technology, Rourkela
520 ## - SUMMARY, ETC.
Summary/Abstract, etc Frequency and Harmonics are two important <br/>parameters for power system control and <br/>protection, power system relaying, power qua<br/>lity monitoring, operation and control of <br/>electrical equipments. Some existing appro<br/>aches of frequency and harmonics estimation <br/>are Fast Fourier Transform (FFT), Least <br/>Square (LS), Least Mean Square (LMS), <br/>Recursive Least Square (RLS), Kalman Filte<br/>ring (KF), Soft Computing Techniques such <br/>as Neural Networks and Ge<br/>netic Algorithms etc. FFT ba<br/>sed technique suffers from <br/>leakage effect i.e. an effect in the frequenc<br/>y analysis of finite length signals and the <br/>performance is highly degraded while estim<br/>ating inter-harmonics and sub-harmonics <br/>including frequency deviations. Recursive estim<br/>ation is not possible in case of LS. LMS <br/>provides poor estimation performance owi<br/>ng to its poor convergence rate as the <br/>adaptation step-size is fixed. In<br/> case of RLS and KF, suitable in<br/>itial choice of covariance <br/>matrix and gain leading to faster convergence<br/> on Mean Square Error is difficult. Initial <br/>choice of Weight vector and learning paramete<br/>r affect the convergence characteristic of <br/>neural estimator. Genetic based algor<br/>ithms takes more time for convergence. <br/>To achieve faster convergence and more accuracy<br/> in estimation, in this thesis a Variable <br/>Leaky Least Mean Square (VL-LMS) is propos<br/>ed for frequency estimation. The proposed <br/>approach uses a variable leak adjustment t<br/>echnique to avoid drifting of the parameters <br/>involved in the estimation mechanism. A <br/>variable adaptation step-size is also <br/>incorporated in the algorithm to yield fa<br/>ster convergence. The performance of the <br/>proposed algorithm is studied through simulatio<br/>ns and on experimental data for several <br/>critical cases such as in presence of <br/>noise, jump in frequency, harmonics and sub-<br/>harmonics and inter-harmonics that often ar<br/>ise in a power system<br/>. These studies show <br/>that the VLLMS algorithm is superior to the existing ones in estimating power system <br/>frequency. <br/>Subsequently, a nonlinear state estimation t<br/>echnique for estimation of harmonics, inter-<br/>harmonics and sub-harmonics based on Ensemb<br/>le Kalman Filtering (EnKF) is proposed. <br/>This algorithm is suitable for problems havi<br/>ng more numbers of variables. In harmonics <br/>estimation problem, there are twelve unknown va<br/>riables for distorted <br/>signal containing <br/>iv<br/>noise and d.c. offsets. EnKFs represent the distribution of power sy<br/>stem states using a <br/>collection of state vectors known as ensemb<br/>le, which replaces the covariance matrix by <br/>sample covariance computed from ensemble. The proposed EnKF estimation technique <br/>accurately estimates the harmonics, sub-ha<br/>rmonics and inter-harmonics including <br/>possible variations in amplitude in the time <br/>domain signal. Further, performance of the <br/>proposed EnKF method is compared with exis<br/>ting techniques such as Least Mean Square <br/>(LMS), Recursive Least Square (RLS), R<br/>ecursive Least Mean Square (RLMS) and <br/>Kalman Filter (KF) algorithms, and it provide<br/>s highly improved results with respect to <br/>tracking time and accuracy. <br/>The thesis also proposed four hybrid estim<br/>ation algorithms such as KF-Adaline, RLS-<br/>Adaline and RLS-BFO (Bacterial Foraging Op<br/>timization), Adaline-BFO, for estimation <br/>of power system frequency and harmonics. <br/>In case of KF-Adaline and RLS-Adaline, <br/>weights of the Adaline are updated using KF<br/> and RLS algorithm(s). Efficacies of the <br/>above hybrid estimation algorithms have been<br/> studied through simulation on numerical <br/>and experimental data that in terms of es<br/>timation accuracy, processing and tracking time, <br/>KF-Adaline outperforms RLS-Adaline. In this<br/> thesis, the proposed hybrid approaches to <br/>power system frequency and harmonics esti<br/>mation first optimize the unknown parameters <br/>of the regressor of the input<br/> power system signal exploiting evolutionary optimization <br/>approach (BFO) and then RLS or Adaline are <br/>applied for achieving faster convergence in <br/>estimating the frequency and harm<br/>onics of distorted signal. <br/>The estimation achieved by application of the proposed estimation approaches are <br/>exploited to design a Hybrid Active Power <br/>Filter (HAPF) for ach<br/>ieving pure sinusoidal <br/>signal. A HAPF has been proposed that us<br/>es modified PWM control technique for <br/>elimination of harmonics in distorted power sy<br/>stem signals. This filter uses the estimation <br/>obtained from KF-Adaline approach. The m<br/>odified PWM control <br/>technique used in <br/>HAPF is based on comparing simultaneously <br/>a triangular high fre<br/>quency carrier signal <br/>with a controlling signal and its 180<br/>0<br/> out of phase signal. A <br/>laboratory prototype for <br/>HAPF with modified PWM control is develope<br/>d for harmonics elimination in a distorted <br/>power system signal arising due<br/> to rectifier load. Simulation and experimentation are <br/>performed to verify the efficacy of the m<br/>odified PWM control based HAPF. The above <br/>designed HAPF exhibits better filtering ability <br/>as compared to passive and active filters.
650 ## - SUBJECT ADDED ENTRY--TOPICAL TERM
Topical Term Engineering and Technology
650 ## - SUBJECT ADDED ENTRY--TOPICAL TERM
Topical Term Electrical Engineering
650 ## - SUBJECT ADDED ENTRY--TOPICAL TERM
Topical Term Power Electronics
700 ## - ADDED ENTRY--PERSONAL NAME
Personal name Subudhi, Bidyadhara
Relator term Supervisor
700 ## - ADDED ENTRY--PERSONAL NAME
Personal name Panda, Abni Mohan
Relator term Supervisor
710 ## - ADDED ENTRY--CORPORATE NAME
Corporate name or jurisdiction name as entry element Department of Electrical Engineering
856 ## - ELECTRONIC LOCATION AND ACCESS
Uniform Resource Identifier http://ethesis.nitrkl.ac.in/2953/
942 ## - ADDED ENTRY ELEMENTS (KOHA)
Koha item type Thesis (Ph.D/M.Tech R)
Holdings
Withdrawn status Damaged status Collection code Permanent Location Current Location Shelving location Date acquired Accession Number Koha item type
    Reference BP Central Library BP Central Library Thesis Section 12/07/2017 T133 Thesis (Ph.D/M.Tech R)

Implemented and Maintained by Biju Patnaik Central Library.
For any Suggestions/Query Contact to library or Email: library@nitrkl.ac.in OR bpcl-cir@nitrkl.ac.in. Ph:91+6612462103
Website/OPAC best viewed in Mozilla Browser in 1366X768 Resolution.

Powered by Koha