KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Search Results

Now showing 1 - 8 of 8

Some of the metrics are blocked by your
consent settings
An asymptotic regression predictive model for sphere gap voltage with gap diameter - And gap spacing
(2006-12-01)
Chatpattananan, V.
This document develops two asymptotic regression models to estimate the value of sphere gap breakdown voltage from any value of gap diameter and gap spacing under IEC 60052 table 2 and table 3 to facilitate those two IEC tables that provide limited values consisting of 12 sizes in sphere diameter ranging from 20 mm to 2000 mm and sphere gap spacing range from 0.5 mm to 1500 mm. In those two IEC tables, the relationship of sphere gap voltages from sphere gap diameter and gap spacing is nonlinear in which asymptotic regression models are proposed is in the form Voltage = b<inf>1</inf> + b<inf>2</inf>e <sup>b3Spacing</sup> + b<inf>4</inf>Diameter that follows the law of diminishing returns that imitates the relationship between sphere gap voltages versus sphere gap diameter and gap spacing. By using this asymptotic regression, this document provides expressions of standard sphere gap voltages as functions of the sphere gap diameter and gap spacing. The expressions are in high accuracy measured by r<sup>2</sup> values. These proposed asymptotic regression equations can be used to estimate sphere gap voltage breakdown from any values of sphere gap diameter and gap spacing that are not specified by IEC 60052 tables. © 2006 IEEE.
Some of the metrics are blocked by your
consent settings
A classification of partial discharge on high voltage equipment with multinomial logistic regression
(2006-12-01)
Chatpattananan, V.
;
Pattanadech, N.
This document proposes a statistical approach in multinomial logistic regression to classify PD patterns into four categories listed as corona at high voltage side in air, corona at low voltage side in air, surface in air, and internal discharge. The independent variables in this multinomial logistic regression model are skewness, kurtosis, asymmetry, and cross correlation following the ø -q -n PD patterns obtained from the fingerprint analysis which is a digital signal processing technique for PD measurement. The experiments were set to simulate all four PD patterns to obtain statistical parameters into 10 independent variables from the fingerprint analysis. This document also applied stepwise model selection technique to reduce from 10 independent variables to 2 independent variables that not only reduces the complexity of the model estimated but also retains the accuracy of this predictive model to 96.2 percent. © 2006 IEEE.
Some of the metrics are blocked by your
consent settings
A linear model for the differences of breakdown voltages in IEC 60052 tables
(2006-12-01)
Chatpattananan, V.
;
Pattanadech, N.
IEC 60052 table 2 and table 3 provides the values of breakdown voltage for sphere gap occurred from different values of gap diameter and gap spacing. This document investigates the breakdown voltages from IEC 60052 table 2 and IEC 60052 table 3 under the same values of sphere gap diameter and gap spacing. The result shows that, provided the same condition of sphere gap diameter and gap spacing, the breakdown voltage value from table 3 (positive impulse) is higher than the breakdown voltage value from table 2 (negative impulse). Also, the relationship between the difference in breakdown voltage with diameter and spacing is simply a linear model in which a least square regression that minimizes sum of square errors can be used to find the relationship. With least square approach, a linear regression equation without an intercept term, or linear regression through the origin, is better accurate than the linear regression with an intercept term. The difference in breakdown voltage also reveals an incongruity in a step function pattern in the data in which an adjustment in removal of some data points was made to propose another regression through the origin equation the explain the difference in breakdown voltage to gap diameter and gap spacing. ©2006 IEEE.
Some of the metrics are blocked by your
consent settings
Linear discriminant analysis for partial discharge classification on high voltage equipment
(2006-12-01)
Chatpattananan, V.
This document is addressing a problem in classifying partial discharge types using linear discriminant analysis. Four PD types generated in the lab are corona at high voltage side in air, corona at low voltage side in air, surface in air, and internal discharge. The independent variables used in this linear discriminant analysis model mainly are from skewness, kurtosis, asymmetry, and cross correlation following the φ - q - n PD patterns obtained from the fingerprint analysis which is a digital signal processing technique for PD measurement. This document also applied stepwise model selection technique to reduce from 10 independent variables to 8 independent variables that not only reduces the complexity of the model estimated but also retains the accuracy of this predictive model to 98.7 percent. © 2006 IEEE.
Some of the metrics are blocked by your
consent settings
Partial discharge classification on high voltage equipment with K-means
(2006-01-01)
Chatpattananan, V.
;
Pattanadech, N.
;
Yutthagowith, P.
This document proposes a statistical classification model using k-means cluster analysis to classify partial discharge (PD) patterns into four categories listed as corona at high voltage side in air, corona at low voltage side in air, surface in air, and internal discharge. The independent variables in this k-means cluster analysis model are skewness, kurtosis, asymmetry, and cross correlation following the φ - q - n PD patterns obtained from the fingerprint analysis which is a digital signal processing technique for PD measurement. The experiments were set to simulate all four PD patterns to obtain statistical parameters into 10 independent variables from the fingerprint analysis. This document recommends using 8 final clusters instead of using 4 final clusters to achieve the accuracy of using this k-means cluster approach to 100 percent. © 2006 IEEE.
Some of the metrics are blocked by your
consent settings
QUEST tree model for partial discharge classification on high voltage equipment
(2006-01-01)
Chatpattananan, V.
;
Pattanadech, N.
;
Vicetjindavat, K.
This document proposes a statistical classification model using QUEST (Quick, Unbiased, and Efficient Statistical Tree) to classify PD patterns into four categories listed as corona at high voltage side in air, corona at low voltage side in air, surface in air, and internal discharge. The independent variables in this QUEST model are skewness, kurtosis, asymmetry, and cross correlation following the φ-q-n PD patterns obtained from the fingerprint analysis which is a digital signal processing technique for PD measurement. The experiments were set to simulate all four PD patterns to obtain statistical parameters into 10 independent variables from the fingerprint analysis. This QUEST classification model can reduce the number of independent variables used to estimate from 10 independent variables to 3 independent variables with the classification accuracy of 100 percent. © 2006 IEEE.
Some of the metrics are blocked by your
consent settings
The effect of convexity of lightning impulse voltage waveforms on accuracy of HV measuring systems
(2006-01-01)
Yutthagowith, P.
;
Pattanadech, N.
;
Chatpattananan, V.
;
Phoomvuthisarn, S.
This paper presents the effect of convexity of lightning impulse voltage waveforms on accuracy of HV measuring systems. The effects of different three convexities on wave front are analyzed by simulation tests with some standard waveforms. The simulation tests have two aspects. First one is consideration in frequency domain by using FFT with standard waveforms and transfer function of measuring system and second one is consideration in time domain by using time convolution method or Duhamel's integration method for computing waveform outputs standard testing waveforms and unit step responses of the sample measuring systems. The standard waveforms under investigations are full lightning impulse voltage waveforms, tail-chopped of lightning impulse voltage waveforms, front-chopped of lightning impulse voltage waveforms according to IEC standards 60060-2. This paper considers two samples of different types of the measuring systems. The results show that the convexity has effect on accuracy of measuring systems. © 2006 IEEE.
Some of the metrics are blocked by your
consent settings
PCA-LDA for partial discharge classification on high voltage equipment
(2006-01-01)
Chatpattananan, V.
;
Pattanadech, N.
;
Vicetjindavat, K.
This document proposes a statistical classification model using principal component analysis (PCA) for a data reduction approach combined with linear discriminant analysis (LDA) for a classification purpose, so called, PCA-LDA to classify PD patterns into four categories listed as corona at high voltage side in air, corona at low voltage side in air, surface in air, and internal discharge. The original independent variables in this classification model are skewness, kurtosis, asymmetry, and cross correlation following the φ-q-n PD patterns obtained from the fingerprint analysis which is a digital signal processing technique for PD measurement. The experiments were set to simulate all four PD patterns to obtain statistical parameters into 10 independent variables from the fingerprint analysis. With PCA, the original 10 independent variables can be reduced to 3 independent variables and using PCA can retains the accuracy of this classification model to 97.5 percent. © 2006 IEEE.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results