Signal Processing and Soft Computing Approaches to Power Signal Frequency and Harmonics Estimation

By: Ray , Pravat KumarContributor(s): Subudhi, Bidyadhara [Supervisor] | Panda, Abni Mohan [Supervisor] | Department of Electrical EngineeringMaterial type: TextTextLanguage: English Publisher: 2011Description: 261 pSubject(s): Engineering and Technology | Electrical Engineering | Power ElectronicsOnline resources: Click here to access online Dissertation note: Thesis (Ph.D)- National Institute of Technology, Rourkela Summary: Frequency and Harmonics are two important parameters for power system control and protection, power system relaying, power qua lity monitoring, operation and control of electrical equipments. Some existing appro aches of frequency and harmonics estimation are Fast Fourier Transform (FFT), Least Square (LS), Least Mean Square (LMS), Recursive Least Square (RLS), Kalman Filte ring (KF), Soft Computing Techniques such as Neural Networks and Ge netic Algorithms etc. FFT ba sed technique suffers from leakage effect i.e. an effect in the frequenc y analysis of finite length signals and the performance is highly degraded while estim ating inter-harmonics and sub-harmonics including frequency deviations. Recursive estim ation is not possible in case of LS. LMS provides poor estimation performance owi ng to its poor convergence rate as the adaptation step-size is fixed. In case of RLS and KF, suitable in itial choice of covariance matrix and gain leading to faster convergence on Mean Square Error is difficult. Initial choice of Weight vector and learning paramete r affect the convergence characteristic of neural estimator. Genetic based algor ithms takes more time for convergence. To achieve faster convergence and more accuracy in estimation, in this thesis a Variable Leaky Least Mean Square (VL-LMS) is propos ed for frequency estimation. The proposed approach uses a variable leak adjustment t echnique to avoid drifting of the parameters involved in the estimation mechanism. A variable adaptation step-size is also incorporated in the algorithm to yield fa ster convergence. The performance of the proposed algorithm is studied through simulatio ns and on experimental data for several critical cases such as in presence of noise, jump in frequency, harmonics and sub- harmonics and inter-harmonics that often ar ise in a power system . These studies show that the VLLMS algorithm is superior to the existing ones in estimating power system frequency. Subsequently, a nonlinear state estimation t echnique for estimation of harmonics, inter- harmonics and sub-harmonics based on Ensemb le Kalman Filtering (EnKF) is proposed. This algorithm is suitable for problems havi ng more numbers of variables. In harmonics estimation problem, there are twelve unknown va riables for distorted signal containing iv noise and d.c. offsets. EnKFs represent the distribution of power sy stem states using a collection of state vectors known as ensemb le, which replaces the covariance matrix by sample covariance computed from ensemble. The proposed EnKF estimation technique accurately estimates the harmonics, sub-ha rmonics and inter-harmonics including possible variations in amplitude in the time domain signal. Further, performance of the proposed EnKF method is compared with exis ting techniques such as Least Mean Square (LMS), Recursive Least Square (RLS), R ecursive Least Mean Square (RLMS) and Kalman Filter (KF) algorithms, and it provide s highly improved results with respect to tracking time and accuracy. The thesis also proposed four hybrid estim ation algorithms such as KF-Adaline, RLS- Adaline and RLS-BFO (Bacterial Foraging Op timization), Adaline-BFO, for estimation of power system frequency and harmonics. In case of KF-Adaline and RLS-Adaline, weights of the Adaline are updated using KF and RLS algorithm(s). Efficacies of the above hybrid estimation algorithms have been studied through simulation on numerical and experimental data that in terms of es timation accuracy, processing and tracking time, KF-Adaline outperforms RLS-Adaline. In this thesis, the proposed hybrid approaches to power system frequency and harmonics esti mation first optimize the unknown parameters of the regressor of the input power system signal exploiting evolutionary optimization approach (BFO) and then RLS or Adaline are applied for achieving faster convergence in estimating the frequency and harm onics of distorted signal. The estimation achieved by application of the proposed estimation approaches are exploited to design a Hybrid Active Power Filter (HAPF) for ach ieving pure sinusoidal signal. A HAPF has been proposed that us es modified PWM control technique for elimination of harmonics in distorted power sy stem signals. This filter uses the estimation obtained from KF-Adaline approach. The m odified PWM control technique used in HAPF is based on comparing simultaneously a triangular high fre quency carrier signal with a controlling signal and its 180 0 out of phase signal. A laboratory prototype for HAPF with modified PWM control is develope d for harmonics elimination in a distorted power system signal arising due to rectifier load. Simulation and experimentation are performed to verify the efficacy of the m odified PWM control based HAPF. The above designed HAPF exhibits better filtering ability as compared to passive and active filters.
Tags from this library: No tags from this library for this title. Log in to add tags.
    Average rating: 0.0 (0 votes)
Item type Current location Collection Call number Status Date due Barcode
Thesis (Ph.D/M.Tech R) Thesis (Ph.D/M.Tech R) BP Central Library
Thesis Section
Reference Not for loan T133

Thesis (Ph.D)- National Institute of Technology, Rourkela

Frequency and Harmonics are two important
parameters for power system control and
protection, power system relaying, power qua
lity monitoring, operation and control of
electrical equipments. Some existing appro
aches of frequency and harmonics estimation
are Fast Fourier Transform (FFT), Least
Square (LS), Least Mean Square (LMS),
Recursive Least Square (RLS), Kalman Filte
ring (KF), Soft Computing Techniques such
as Neural Networks and Ge
netic Algorithms etc. FFT ba
sed technique suffers from
leakage effect i.e. an effect in the frequenc
y analysis of finite length signals and the
performance is highly degraded while estim
ating inter-harmonics and sub-harmonics
including frequency deviations. Recursive estim
ation is not possible in case of LS. LMS
provides poor estimation performance owi
ng to its poor convergence rate as the
adaptation step-size is fixed. In
case of RLS and KF, suitable in
itial choice of covariance
matrix and gain leading to faster convergence
on Mean Square Error is difficult. Initial
choice of Weight vector and learning paramete
r affect the convergence characteristic of
neural estimator. Genetic based algor
ithms takes more time for convergence.
To achieve faster convergence and more accuracy
in estimation, in this thesis a Variable
Leaky Least Mean Square (VL-LMS) is propos
ed for frequency estimation. The proposed
approach uses a variable leak adjustment t
echnique to avoid drifting of the parameters
involved in the estimation mechanism. A
variable adaptation step-size is also
incorporated in the algorithm to yield fa
ster convergence. The performance of the
proposed algorithm is studied through simulatio
ns and on experimental data for several
critical cases such as in presence of
noise, jump in frequency, harmonics and sub-
harmonics and inter-harmonics that often ar
ise in a power system
. These studies show
that the VLLMS algorithm is superior to the existing ones in estimating power system
frequency.
Subsequently, a nonlinear state estimation t
echnique for estimation of harmonics, inter-
harmonics and sub-harmonics based on Ensemb
le Kalman Filtering (EnKF) is proposed.
This algorithm is suitable for problems havi
ng more numbers of variables. In harmonics
estimation problem, there are twelve unknown va
riables for distorted
signal containing
iv
noise and d.c. offsets. EnKFs represent the distribution of power sy
stem states using a
collection of state vectors known as ensemb
le, which replaces the covariance matrix by
sample covariance computed from ensemble. The proposed EnKF estimation technique
accurately estimates the harmonics, sub-ha
rmonics and inter-harmonics including
possible variations in amplitude in the time
domain signal. Further, performance of the
proposed EnKF method is compared with exis
ting techniques such as Least Mean Square
(LMS), Recursive Least Square (RLS), R
ecursive Least Mean Square (RLMS) and
Kalman Filter (KF) algorithms, and it provide
s highly improved results with respect to
tracking time and accuracy.
The thesis also proposed four hybrid estim
ation algorithms such as KF-Adaline, RLS-
Adaline and RLS-BFO (Bacterial Foraging Op
timization), Adaline-BFO, for estimation
of power system frequency and harmonics.
In case of KF-Adaline and RLS-Adaline,
weights of the Adaline are updated using KF
and RLS algorithm(s). Efficacies of the
above hybrid estimation algorithms have been
studied through simulation on numerical
and experimental data that in terms of es
timation accuracy, processing and tracking time,
KF-Adaline outperforms RLS-Adaline. In this
thesis, the proposed hybrid approaches to
power system frequency and harmonics esti
mation first optimize the unknown parameters
of the regressor of the input
power system signal exploiting evolutionary optimization
approach (BFO) and then RLS or Adaline are
applied for achieving faster convergence in
estimating the frequency and harm
onics of distorted signal.
The estimation achieved by application of the proposed estimation approaches are
exploited to design a Hybrid Active Power
Filter (HAPF) for ach
ieving pure sinusoidal
signal. A HAPF has been proposed that us
es modified PWM control technique for
elimination of harmonics in distorted power sy
stem signals. This filter uses the estimation
obtained from KF-Adaline approach. The m
odified PWM control
technique used in
HAPF is based on comparing simultaneously
a triangular high fre
quency carrier signal
with a controlling signal and its 180
0
out of phase signal. A
laboratory prototype for
HAPF with modified PWM control is develope
d for harmonics elimination in a distorted
power system signal arising due
to rectifier load. Simulation and experimentation are
performed to verify the efficacy of the m
odified PWM control based HAPF. The above
designed HAPF exhibits better filtering ability
as compared to passive and active filters.

There are no comments on this title.

to post a comment.

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