Electrical and Computer Systems Engineering Postgraduate Student Research Forum 2001 WAVELET ANALYSIS FOR CONDITION MONITORING OF CIRCUIT BREAKERS Author:

Slides:



Advertisements
Similar presentations
An Overview of ABFT in cloud computing
Advertisements

Introduction to Switchgear
Visualization of dynamic power and synchrony changes in high density EEG A. Alba 1, T. Harmony2, J.L. Marroquín 2, E. Arce 1 1 Facultad de Ciencias, UASLP.
POSTER TEMPLATE BY: Multi-Sensor Health Diagnosis Using Deep Belief Network Based State Classification Prasanna Tamilselvan.
Introduction The aim the project is to analyse non real time EEG (Electroencephalogram) signal using different mathematical models in Matlab to predict.
APPLICATION OF NEURAL NETWORKS AND WAVELET TRANSFORMS IN HIGH IMPEDANCE FAULT DETECTION IN ELECTRICAL SYSTEMS A. M. Sharaf, SMIEEE S. M. A. Saleem Department.
SIGNAL PROCESSING TECHNIQUES USED FOR THE ANALYSIS OF ACOUSTIC SIGNALS FROM HEART AND LUNGS TO DETECT PULMONARY EDEMA 1 Pratibha Sharma Electrical, Computer.
Visualizing Heart Data from Pulse Intervals By Juan Gabriel Estrada Alvarez.
CONTENT BASED FACE RECOGNITION Ankur Jain 01D05007 Pranshu Sharma Prashant Baronia 01D05005 Swapnil Zarekar 01D05001 Under the guidance of Prof.
Applications of Wavelet Transform and Artificial Neural Network in Digital Signal Detection for Indoor Optical Wireless Communication Sujan Rajbhandari.
ExaSphere Network Analysis Engine © 2006 Joseph E. Johnson, PhD
Multi-Scale Analysis for Network Traffic Prediction and Anomaly Detection Ling Huang Joint work with Anthony Joseph and Nina Taft January, 2005.
1 Using A Multiscale Approach to Characterize Workload Dynamics Characterize Workload Dynamics Tao Li June 4, 2005 Dept. of Electrical.
School of Computing and Engineering Diagnostic Engineering Research Group Diagnosis and Prognosis of Machinery Health based on Advanced Intelligent Computations.
SIGDIG – Signal Discrimination for Condition Monitoring A system for condition analysis and monitoring of industrial signals Collaborative research effort.
Smart Traveller with Visual Translator for OCR and Face Recognition LYU0203 FYP.
RACE: Time Series Compression with Rate Adaptivity and Error Bound for Sensor Networks Huamin Chen, Jian Li, and Prasant Mohapatra Presenter: Jian Li.
Wind turbine induction generator bearing fault detection using stator current analysis By School of Electrical and Electronic Engineering The University.
Discrete Time Periodic Signals A discrete time signal x[n] is periodic with period N if and only if for all n. Definition: Meaning: a periodic signal keeps.
SOMTIME: AN ARTIFICIAL NEURAL NETWORK FOR TOPOLOGICAL AND TEMPORAL CORRELATION FOR SPATIOTEMPORAL PATTERN LEARNING.
DIGITAL SIGNAL PROCESSING IN ANALYSIS OF BIOMEDICAL IMAGES Prof. Aleš Procházka Institute of Chemical Technology in Prague Department of Computing and.
EE513 Audio Signals and Systems Statistical Pattern Classification Kevin D. Donohue Electrical and Computer Engineering University of Kentucky.
IIS for Image Processing Michael J. Watts
ERP DATA ACQUISITION & PREPROCESSING EEG Acquisition: 256 scalp sites; vertex recording reference (Geodesic Sensor Net)..01 Hz to 100 Hz analogue filter;
Fault Prediction in Electrical Valves Using Temporal Kohonen Maps Luiz F. Gonçalves, Eduardo L. Schneider, Jefferson L. Bosa, Renato Ventura B. Henriques,
Presented by Tienwei Tsai July, 2005
ea technology Effective Condition Assessment of MV Switchgear
Robust Fault analysis Technique for Permanent Magnet DC Motor In safety Critical Applications Wathiq Abed Wathiq Abed Supervisor - Sanjay Sharma University.
Element 2: Discuss basic computational intelligence methods.
1 UFRGS Design of an Embedded System for the Proactive Maintenance of Electrical Valves Luiz F. Gonçalves, Jefferson L. Bosa, Renato V. B. Henriques, Marcelo.
Electrical and Computer Systems Engineering Postgraduate Student Research Forum 2001 Experimental measurements of dielectric and conduction properties.
WAVELET (Article Presentation) by : Tilottama Goswami Sources:
COMPARISON OF IMAGE ANALYSIS FOR THAI HANDWRITTEN CHARACTER RECOGNITION Olarik Surinta, chatklaw Jareanpon Department of Management Information System.
RECENT DEVELOPMENTS OF INDUCTION MOTOR DRIVES FAULT DIAGNOSIS USING AI TECHNIQUES 1 Oly Paz.
Deriving connectivity patterns in the primary visual cortex from spontaneous neuronal activity and feature maps Barak Blumenfeld, Dmitri Bibitchkov, Shmuel.
RIDA: A Robust Information-Driven Data Compression Architecture for Irregular Wireless Sensor Networks Nirupama Bulusu (joint work with Thanh Dang, Wu-chi.
Handwritten Recognition with Neural Network Chatklaw Jareanpon, Olarik Surinta Mahasarakham University.
Univ logo Fault Diagnosis for Power Transmission Line using Statistical Methods Yuanjun Guo Prof. Kang Li Queen’s University, Belfast UKACC PhD Presentation.
©2009 Mladen Kezunovic. Improving Relay Performance By Off-line and On-line Evaluation Mladen Kezunovic Jinfeng Ren, Chengzong Pang Texas A&M University,
A New Methodology for Systematic Conceptual Design by means of Generalized Discrete Representations Research group conducted by Dr. Offer Shai Department.
SOM-based Data Visualization Methods Author:Juha Vesanto Advisor:Dr. Hsu Graduate:ZenJohn Huang IDSL seminar 2002/01/24.
Wavelets Anderson G Moura 05/29/2015. Introduction Biomedical signals usually consist of brief high-frequency components closely spaced in time, accompanied.
Figure 3. Log-log plot of simulated oscillating phantom, assuming a Gaussian-shaped field. Field constants a 1 =a 2 =0.1. The data initially plateau, then.
J.-Y. Yang, J.-S. Wang and Y.-P. Chena, Using acceleration measurements for activity recognition: An effective learning algorithm for constructing neural.
Wavelets Pedro H. R. Garrit 05/209/2015.
Speech Lab, ECE, State University of New York at Binghamton  Classification accuracies of neural network (left) and MXL (right) classifiers with various.
MSc Project Musical Instrument Identification System MIIS Xiang LI ee05m216 Supervisor: Mark Plumbley.
Approach Design Definition with Synthesis and Analysis of Alternative Solutions Leading to a Decision.
Neural Network Application for Fault Analysis
CLASSIFICATION OF ECG SIGNAL USING WAVELET ANALYSIS
KILOMETRIC FAULTS: NATURE, AFFECTING PARAMETERS, AND IMPACT ON THE BREAKER STRESSES _____________________________ Mohamed M. Saied Electrical Engineering.
National Taiwan Normal A System to Detect Complex Motion of Nearby Vehicles on Freeways C. Y. Fang Department of Information.
SWEEP FREQUENCY RESPONSE ANALYSIS AS A DIANOSTIC TOOL TO DETECT TRANSFORMER MECHANICAL INTEGRITY by:Luwendran Moodley Brian de Klerk ETHEKWINI ELECTRICITY.
Korea Advanced Institute of Science and Technology JUL 5th, 2010 Seung Min Lee 1.
Electronics And Communications Engineering Nalla Malla Reddy Engineering College Major Project Seminar on “Phase Preserving Denoising of Images” Guide.
Extraction of surface impedance from magnetotelluric data
Topic: Waveforms in Noesis
Data Transformation: Normalization
Development of RCP Vibration Monitoring System using Power Line Analysis Method 정 재 천.
ARTIFICIAL NEURAL NETWORKS
Wavelets : Introduction and Examples
IIS for Image Processing
Multiple Wavelet Coefficients Fusion in Deep Residual Networks for Fault Diagnosis
WiFinger: Talk to Your Smart Devices with Finger-grained Gesture
DDoS Attack Detection under SDN Context
National Conference on Recent Advances in Wireless Communication & Artificial Intelligence (RAWCAI-2014) Organized by Department of Electronics & Communication.
EE513 Audio Signals and Systems
LOAD BEHAVIOUR DURING VOLTAGE DIPS
Neural Network Pipeline CONTACT & ACKNOWLEDGEMENTS
Achintya Choudhury Bhartiya Skill Development University Jaipur India
Presentation transcript:

Electrical and Computer Systems Engineering Postgraduate Student Research Forum 2001 WAVELET ANALYSIS FOR CONDITION MONITORING OF CIRCUIT BREAKERS Author: Dennis Lee CEPE Electrical Power Engineering Supervisors: Prof. RE Morrison Brian Lithgow INTRODUCTION Most of the major failures in a circuit breaker are of mechanical origin. Effective diagnosis of any early deterioration of the mechanical components is important to maintain the reliability of the circuit breaker. To achieve this, vibration analysis is needed. Figure 1: Vibration Monitoring of 66kV Circuit Breakers. The signals are recorded in a data logger. Figure 2: Faulty vibration time series and FFT for closing operation of the circuit breaker. The time series are made of many singularities that contain important details of the mechanical condition of the circuit breaker. HYPOTHESIS Mal-adjustments of circuit breaker’s components will alter the vibration time-frequencies of the circuit breaker during switching (Fig. 1). By comparing the vibration time- frequencies with a reference, usually healthy samples, the potential faults can be detected and identified (Fig. 2). Legends: X: Wavelet features for 4 measurement points for 5 opening operations of a simulated fault on the circuit breaker. O: Wavelet features for the healthy circuit breaker. Figure 3: A wavelet condition map shows a series of condition vectors derived from 20 tests. This map is a visual representation of the condition of the circuit breaker represented as a point in an Argand plane. METHODOLOGY The Wavelet Packet Transform is originally applied to the fault diagnosis of circuit breakers and the algorithm is tested using the real data collected. The wavelet maxima are used as feature descriptors to plot a condition map (Fig. 3). The 2 prominent wavelet coefficients correspond to resonant frequencies within the mechanical system of the circuit breaker. These features are then input into neural networks for classification. DISCUSSIONS & CONCLUSIONS The results in the wavelet based condition map show that faults on circuit breakers can be detected as the healthy and faulty features are separate into different clusters. These clusters are successfully classified using neural networks. The new condition map identifies clearly the vibration features due to the fault. This is because the vibration time series consists of a few “spikes” which are regular in patern. This type of pattern is effectively identified using the wavelet analysis. EXPERIMENTAL RESULTS A. Detection of Fault types 1. Opening sample size detection failures success rate B (contact penetration +) D (contact penetration -) F (tail spring compression +) H (tail spring compression -) Closing B D F H standard deviation = average success rate = 93.02%; average success rate of B& D = 92.03% average success rate of F & H = 94.01% confidence interval of success rate =  correlation factors between opening and closing data = correlation factors between faults B D & F H = confidence level is 5%