Analysis of distorted waveforms using parametric spectrum estimation methods and robust averaging Zbigniew LEONOWICZ 13 th Workshop on High Voltage Engineering.

Slides:



Advertisements
Similar presentations
Chapter 3 Properties of Random Variables
Advertisements

I OWA S TATE U NIVERSITY Department of Animal Science Using Basic Graphical and Statistical Procedures (Chapter in the 8 Little SAS Book) Animal Science.
Principal Component Analysis Based on L1-Norm Maximization Nojun Kwak IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008.
Probability Distributions CSLU 2850.Lo1 Spring 2008 Cameron McInally Fordham University May contain work from the Creative Commons.
© 2008 McGraw-Hill Higher Education The Statistical Imagination Chapter 4. Measuring Averages.
Chap 10: Summarizing Data 10.1: INTRO: Univariate/multivariate data (random samples or batches) can be described using procedures to reveal their structures.
STAT 497 APPLIED TIME SERIES ANALYSIS
Cox Model With Intermitten and Error-Prone Covariate Observation Yury Gubman PhD thesis in Statistics Supervisors: Prof. David Zucker, Prof. Orly Manor.
Calculating & Reporting Healthcare Statistics
Descriptive Statistics Statistical Notation Measures of Central Tendency Measures of Variability Estimating Population Values.
A quick introduction to the analysis of questionnaire data John Richardson.
Lecture 4 Measurement Accuracy and Statistical Variation.
Edpsy 511 Homework 1: Due 2/6.
Overview of Robust Methods Analysis Jinxia Ma November 7, 2013.
 Deviation is a measure of difference for interval and ratio variables between the observed value and the mean.  The sign of deviation (positive or.
Statistics Introduction 1.)All measurements contain random error  results always have some uncertainty 2.)Uncertainty are used to determine if two or.
July, 2000Guang Jin Statistics in Applied Science and Technology Chapter 4 Summarizing Data.
Zbigniew LEONOWICZ, Tadeusz LOBOS Wroclaw University of Technology Wroclaw University of Technology, Poland International Conference.
1 Summarizing Performance Data Confidence Intervals Important Easy to Difficult Warning: some mathematical content.
Mean Tests & X 2 Parametric vs Nonparametric Errors Selection of a Statistical Test SW242.
Statistical Methods For Engineers ChE 477 (UO Lab) Larry Baxter & Stan Harding Brigham Young University.
Chapter 3: Central Tendency. Central Tendency In general terms, central tendency is a statistical measure that determines a single value that accurately.
Descriptive Statistics Anwar Ahmad. Central Tendency- Measure of location Measures descriptive of a typical or representative value in a group of observations.
by B. Zadrozny and C. Elkan
Quantitative Skills: Data Analysis
© 2006 McGraw-Hill Higher Education. All rights reserved. Numbers Numbers mean different things in different situations. Consider three answers that appear.
University of Ottawa - Bio 4118 – Applied Biostatistics © Antoine Morin and Scott Findlay 08/10/ :23 PM 1 Some basic statistical concepts, statistics.
Why statisticians were created Measure of dispersion FETP India.
Thinking About Psychology: The Science of Mind and Behavior 2e Charles T. Blair-Broeker Randal M. Ernst.
Two Brain Signal (EEG) processing applications Zbigniew Zbigniew LEONOWICZ, PhD Robust estimators & Blind Signal Separation (BSS)
LECTURER PROF.Dr. DEMIR BAYKA AUTOMOTIVE ENGINEERING LABORATORY I.
Central Tendency Introduction to Statistics Chapter 3 Sep 1, 2009 Class #3.
© 2006 McGraw-Hill Higher Education. All rights reserved. Numbers Numbers mean different things in different situations. Consider three answers that appear.
Biostatistics Class 1 1/25/2000 Introduction Descriptive Statistics.
Lecture 2 Forestry 3218 Lecture 2 Statistical Methods Avery and Burkhart, Chapter 2 Forest Mensuration II Avery and Burkhart, Chapter 2.
Psychology’s Statistics. Statistics Are a means to make data more meaningful Provide a method of organizing information so that it can be understood.
Copyright  2003 by Dr. Gallimore, Wright State University Department of Biomedical, Industrial Engineering & Human Factors Engineering Human Factors Research.
Chapter 7 Probability and Samples: The Distribution of Sample Means.
Chapter 3 For Explaining Psychological Statistics, 4th ed. by B. Cohen 1 Chapter 3: Measures of Central Tendency and Variability Imagine that a researcher.
FREQUANCY DISTRIBUTION 8, 24, 18, 5, 6, 12, 4, 3, 3, 2, 3, 23, 9, 18, 16, 1, 2, 3, 5, 11, 13, 15, 9, 11, 11, 7, 10, 6, 5, 16, 20, 4, 3, 3, 3, 10, 3, 2,
NEW POWER QUALITY INDICES Zbigniew LEONOWICZ Department of Electrical Engineering Wroclaw University of Technology, Poland The Seventh IASTED International.
Relative Values. Statistical Terms n Mean:  the average of the data  sensitive to outlying data n Median:  the middle of the data  not sensitive to.
Experiments on Noise CharacterizationRoma, March 10,1999Andrea Viceré Experiments on Noise Analysis l Need of noise characterization for  Monitoring the.
Robust Estimators.
Z bigniew Leonowicz, Wroclaw University of Technology Z bigniew Leonowicz, Wroclaw University of Technology, Poland XXIX  IC-SPETO.
Chapter 3: Central Tendency. Central Tendency In general terms, central tendency is a statistical measure that determines a single value that accurately.
Edpsy 511 Exploratory Data Analysis Homework 1: Due 9/19.
Introduction to statistics I Sophia King Rm. P24 HWB
CHAPTER 2: Basic Summary Statistics
Analysis of Traction System Time-Varying Signals using ESPRIT Subspace Spectrum Estimation Method Z. Leonowicz, T. Lobos
Educational Research: Data analysis and interpretation – 1 Descriptive statistics EDU 8603 Educational Research Richard M. Jacobs, OSA, Ph.D.
Lecture 8: Measurement Errors 1. Objectives List some sources of measurement errors. Classify measurement errors into systematic and random errors. Study.
Two-Sample-Means-1 Two Independent Populations (Chapter 6) Develop a confidence interval for the difference in means between two independent normal populations.
Psychology’s Statistics Appendix. Statistics Are a means to make data more meaningful Provide a method of organizing information so that it can be understood.
Methods of Presenting and Interpreting Information Class 9.
Estimating standard error using bootstrap
Descriptive and Inferential Statistics
Confidence Intervals Cont.
Statistics Use of mathematics to ORGANIZE, SUMMARIZE and INTERPRET numerical data. Needed to help psychologists draw conclusions.
Statistics in Management
ESTIMATION.
MEASUREMENT OF IEC GROUPS AND SUBGROUPS USING ADVANCED SPECTRUM ESTIMATION METHODS A. Bracale, G. Carpinelli, Z. Leonowicz, T. Lobos, J. Rezmer.
Relative Values.
Outlier Processing via L1-Principal Subspaces
Analyzing Redistribution Matrix with Wavelet
Numerical Measures: Centrality and Variability
Science of Psychology AP Psychology
Summary descriptive statistics: means and standard deviations:
Summary descriptive statistics: means and standard deviations:
CHAPTER 2: Basic Summary Statistics
Presentation transcript:

Analysis of distorted waveforms using parametric spectrum estimation methods and robust averaging Zbigniew LEONOWICZ 13 th Workshop on High Voltage Engineering Söllerhaus Austria

Robust averaging Averaging is probably the most widely used basic statistical procedure in experimental science.Averaging is probably the most widely used basic statistical procedure in experimental science. Estimation of the location of data („central tendency”) in the presence of random variations among the observationsEstimation of the location of data („central tendency”) in the presence of random variations among the observations Data variations can be a result of variations in the phenomenon of interest or of some unavoidable measuring errors.Data variations can be a result of variations in the phenomenon of interest or of some unavoidable measuring errors. In signal processing terms, this can be considered as contamination of useful „signal” by useless „noise” linearly added to it.In signal processing terms, this can be considered as contamination of useful „signal” by useless „noise” linearly added to it. Since the noise usually has zero mean, averaging minimizes its contribution, while the signal is preserved, and the signal to noise ratio is improvedSince the noise usually has zero mean, averaging minimizes its contribution, while the signal is preserved, and the signal to noise ratio is improved

Synchronization Averaging consists of applying of any statistical procedure to extract the useful information from the background noise.Averaging consists of applying of any statistical procedure to extract the useful information from the background noise. When useful data are time-locked to some event and the noise is not time-locked, it allows the cancellation of the noise by simple point-by- point data summation.When useful data are time-locked to some event and the noise is not time-locked, it allows the cancellation of the noise by simple point-by- point data summation. This procedure is equivalent to the use of the arithmetic meanThis procedure is equivalent to the use of the arithmetic mean

Review of robust avearging methods Sensitivity of an estimator to the presence of outliers (i.e. data points that deviate from the pattern set by the majority of the data set)Sensitivity of an estimator to the presence of outliers (i.e. data points that deviate from the pattern set by the majority of the data set) Robustness of an estimator is measured by the breakdown valueRobustness of an estimator is measured by the breakdown value How many data points need to be replaced by arbitrary values in order to make the estimator explode (tend to infinity) or implode (tend to zero) ?How many data points need to be replaced by arbitrary values in order to make the estimator explode (tend to infinity) or implode (tend to zero) ? Arithmetic mean has 0% breakdownArithmetic mean has 0% breakdown Median is very robust with breakdown value 50%Median is very robust with breakdown value 50%

Robust location estimators Many location estimators can be presented in unified way by ordering the values of the sample asMany location estimators can be presented in unified way by ordering the values of the sample as and then applying the weight function and then applying the weight function where is a function designed to reduce the influence of certain observations (data points) in form of weighting and represents ordered data.where is a function designed to reduce the influence of certain observations (data points) in form of weighting and represents ordered data.

Examples MedianMedian When the data have the size of (2M+1), the median is the value of the (M +1) th ordered observation. Trimmed meanTrimmed mean For the  -trimmed mean (where p =  N) the weights can be defined as: p highest and p lowest samples are removed.

Winsorized mean Winsorized mean replaces each observation in each  fraction (p =  N) of the tail of the distribution by the value of the nearest unaffected observation.Winsorized mean replaces each observation in each  fraction (p =  N) of the tail of the distribution by the value of the nearest unaffected observation. 0  p  0,25N usually, depending on the heaviness of the tails of the distribution. 0  p  0,25N usually, depending on the heaviness of the tails of the distribution.

Weight functions

Weight functions - other TL-mean applies higher weights for the middle observationsTL-mean applies higher weights for the middle observations tanh estimator applies smoothly changing weights to the values close to extreme, it can be set to ignore extreme valuestanh estimator applies smoothly changing weights to the values close to extreme, it can be set to ignore extreme values

Comparison

Investigations IEC harmonic and interharmonic subgroups calculation IEC Std , IEC harmonic and interharmonic subgroups calculation IEC Std , DFT with 5 Hz resolution in frequency characterize the waveform distortionsDFT with 5 Hz resolution in frequency characterize the waveform distortions

Parametric methods MUSICMUSIC Eigenvalues of the correlation matrix which correspond to the noise subspace used for parameter estimation ESPRITESPRIT based on naturally existing shift invariance between the discrete time series, which leads to rotational invariance between the corresponding signal subspaces. Uses signal subspace.

Progr. average of harmonic groups dc arc furnace supplydc arc furnace supply 11th harmonic group11th harmonic group 2nd interharmonic group2nd interharmonic group

Results MSE Method MSE groups MSE subgroups DFT ESPRIT MUSIC

Advantage of Winsorized mean When comparing values of power quality indices obtained from different parts of the same recorded waveform, a high variability of results appears. To alleviate this problem, winsorized mean was appplied to compute averages from spectral data. When using the value of a=0.2 which means that 20% of ordered data points were discarded and replaced by nearest unaffected data.When comparing values of power quality indices obtained from different parts of the same recorded waveform, a high variability of results appears. To alleviate this problem, winsorized mean was appplied to compute averages from spectral data. When using the value of a=0.2 which means that 20% of ordered data points were discarded and replaced by nearest unaffected data. In such way the outliers were removed and replaced by data, which are assumed to belong to “true” spectral content of investigated waveform.In such way the outliers were removed and replaced by data, which are assumed to belong to “true” spectral content of investigated waveform. The use of winsorized mean instead of usual arithmetic mean allowed reducing the variance of results by nearly 35%.The use of winsorized mean instead of usual arithmetic mean allowed reducing the variance of results by nearly 35%.

Conclusions Results show that the highest improvement of accuracy can be obtained by using the ESPRIT method (especially for interharmonics estimation), closely followed by MUSIC method, which outperform classical DFT approach by over 50%.Results show that the highest improvement of accuracy can be obtained by using the ESPRIT method (especially for interharmonics estimation), closely followed by MUSIC method, which outperform classical DFT approach by over 50%. Partially stochastic nature of investigated arc furnace waveforms caused high variability of calculated power quality indices. The use of robust averaging (winsorized mean) helped to reduce this unwanted variability.Partially stochastic nature of investigated arc furnace waveforms caused high variability of calculated power quality indices. The use of robust averaging (winsorized mean) helped to reduce this unwanted variability.

Conclusions Trimmed estimators are a class of robust estimators of data locations which can help to improve averaging of experimental data when: number of experiments is small data are highly nonstationary data include outliers. Their advantages can be understood as a reasonable compromise between median which is very robust but discard too much information and arithmetic mean conventionally used for averaging which use all data but, due of this, is sensitive to outliers. Additional improvement of averaging can be gained by introducing advanced weighting of ordered data