Fault Diagnosis System Identification Based on Impedance Matching Balance Transformer

Similar documents
Pulse Train Controlled PCCM Buck-Boost Converter Ming Qina, Fangfang Lib

Memorandum on Impulse Winding Tester

Transformer of tgδ on MSP430F1331 single chip microcomputer WANG Han 1 CAI Xinjing 1,XiaoJieping 2,Liu weiqing 2

Phase-Shifting Control of Double Pulse in Harmonic Elimination Wei Peng1, a*, Junhong Zhang1, Jianxin gao1, b, Guangyi Li1, c

EE 330 Lecture 24. Amplification with Transistor Circuits Small Signal Modelling

GG6005. General Description. Features. Applications DIP-8A Primary Side Control SMPS with Integrated MOSFET

Development of Temporary Ground Wire Detection Device

Primary Side Control SMPS with Integrated MOSFET

P. Bruschi: Project guidelines PSM Project guidelines.

Dead Zone Compensation Method of H-Bridge Inverter Series Structure

A1 K. 12V rms. 230V rms. 2 Full Wave Rectifier. Fig. 2.1: FWR with Transformer. Fig. 2.2: Transformer. Aim: To Design and setup a full wave rectifier.

4 20mA Interface-IC AM462 for industrial µ-processor applications

Parameters Affecting Lightning Backflash Over Pattern at 132kV Double Circuit Transmission Lines

Study on the Wide Gap Dielectric Barrier Discharge Device Gaofeng Wang

AN303 APPLICATION NOTE

EE201 Circuit Theory I Fall

White paper. RC223 (type B) residual-current release

THE OSCILLOSCOPE AND NOISE. Objectives:

Investigation and Simulation Model Results of High Density Wireless Power Harvesting and Transfer Method

Monitoring Relays. Current Monitoring A A DIN. Comat products comply with different international standards and are certified accordingly.

A New Voltage Sag and Swell Compensator Switched by Hysteresis Voltage Control Method

Experiment 6: Transmission Line Pulse Response

EXPERIMENT #9 FIBER OPTIC COMMUNICATIONS LINK

Control and Protection Strategies for Matrix Converters. Control and Protection Strategies for Matrix Converters

Foreign Fiber Image Segmentation Based on Maximum Entropy and Genetic Algorithm

Laplacian Mixture Modeling for Overcomplete Mixing Matrix in Wavelet Packet Domain by Adaptive EM-type Algorithm and Comparisons

Direct Analysis of Wave Digital Network of Microstrip Structure with Step Discontinuities

10. The Series Resistor and Inductor Circuit

EXPERIMENT #4 AM MODULATOR AND POWER AMPLIFIER

ORDER INFORMATION TO pin 320 ~ 340mV AMC7150DLF

Motion-blurred star image acquisition and restoration method based on the separable kernel Honglin Yuana, Fan Lib and Tao Yuc

General data 8/91. Overview

A Segmentation Method for Uneven Illumination Particle Images

Application Note 5324

Increasing Measurement Accuracy via Corrective Filtering in Digital Signal Processing

Comparison of ATP Simulation and Microprocessor

Dimensions. Model Number. Electrical connection emitter. Features. Electrical connection receiver. Product information. Indicators/operating means

Electronic timer CT-MVS.12 Multifunctional with 1 c/o contact Data sheet

BOUNCER CIRCUIT FOR A 120 MW/370 KV SOLID STATE MODULATOR

Knowledge Transfer in Semi-automatic Image Interpretation

A New Measurement Method of the Dynamic Contact Resistance of HV Circuit Breakers

Dimensions. Transmitter Receiver ø2.6. Electrical connection. Transmitter +UB 0 V. Emitter selection. = Light on = Dark on

MAP-AIDED POSITIONING SYSTEM

Explanation of Maximum Ratings and Characteristics for Thyristors

Dimensions. Transmitter Receiver ø2.6. Electrical connection. Transmitter +UB 0 V. Emitter selection. = Light on = Dark on

AK8777B. Overview. Features

Automatic Power Factor Control Using Pic Microcontroller

Proceedings of International Conference on Mechanical, Electrical and Medical Intelligent System 2017

GaN-HEMT Dynamic ON-state Resistance characterisation and Modelling

Programmable DC Electronic Load 8600 Series

Universal microprocessor-based ON/OFF and P programmable controller MS8122A MS8122B

Control circuit for a Self-Oscillating Power Supply (SOPS) TDA8385

Communication Systems. Department of Electronics and Electrical Engineering

COMBITRON Program Schedule

Programmable DC Electronic Loads 8600 Series

= f 8 f 2 L C. i C. 8 f C. Q1 open Q2 close (1+D)T DT 2. i C = i L. Figure 2: Typical Waveforms of a Step-Down Converter.

4.5 Biasing in BJT Amplifier Circuits

Chapter 2 Introduction: From Phase-Locked Loop to Costas Loop

Evaluation of Instantaneous Reliability Measures for a Gradual Deteriorating System

Connection. Input II EEx ia IIC without SC red. Composition

MODELING OF CROSS-REGULATION IN MULTIPLE-OUTPUT FLYBACK CONVERTERS

CURRENT MODE PWM+PFM CONTROLLER WITH BUILT-IN HIGH VOLTAGE MOSFET

Application Note AN-1083

Simulation Analysis of DC-DC Circuit Based on Simulink in Intelligent Vehicle Terminal

Power Efficient Battery Charger by Using Constant Current/Constant Voltage Controller

Surviving a strike. ABB helps distribution transformers survive rapid voltage transients

A Novel Bidirectional DC-DC Converter with Battery Protection

Accurate Tunable-Gain 1/x Circuit Using Capacitor Charging Scheme

7 th International Conference on DEVELOPMENT AND APPLICATION SYSTEMS S u c e a v a, R o m a n i a, M a y 27 29,

Table of Contents. 3.0 SMPS Topologies. For Further Research. 3.1 Basic Components. 3.2 Buck (Step Down) 3.3 Boost (Step Up) 3.4 Inverter (Buck/Boost)

Power Loss Research on IGCT-applied NPC Three-level Converter

Photo-electric detectors

PROFET BTS 736 L2. Smart High-Side Power Switch Two Channels: 2 x 40mΩ Status Feedback


Role of Kalman Filters in Probabilistic Algorithm

SLAM Algorithm for 2D Object Trajectory Tracking based on RFID Passive Tags

AN5028 Application note

EE 40 Final Project Basic Circuit

Model Number Structure

Special Insert. Everything you wanted to know about Type B residual current circuit breakers but never dared to ask

16.5 ADDITIONAL EXAMPLES

Using Box-Jenkins Models to Forecast Mobile Cellular Subscription

Programmable DC Electronic Loads 8600 Series

Evaluation of the Digital images of Penaeid Prawns Species Using Canny Edge Detection and Otsu Thresholding Segmentation

Analysis ofthe Effects ofduty Cycle Constraints in Multiple-Input Converters for Photovoltaic Applications

Estimating Transfer Functions with SigLab

A Bidirectional Three-Phase Push-Pull Converter With Dual Asymmetrical PWM Method

Microwave Transistor Oscillator Design

MX6895BETR. -550V Full Bridge Gate Driver INTEGRATED CIRCUITS DIVISION. Features. Description. Applications. Ordering Information

Monitoring Technique. VARIMETER Current Relay BA 9053, MK 9053N

Family of Single-Inductor Multi-Output DC-DC Converters

Switch Basics and Applications. Selection Service Speed

Chapter 2 Summary: Continuous-Wave Modulation. Belkacem Derras

Obsolete Product(s) - Obsolete Product(s)

Optimal configuration algorithm of a satellite transponder

DISCONTINUED MODEL Replaced with Model JPS3

A Control Technique for 120Hz DC Output Ripple-Voltage Suppression Using BIFRED with a Small-Sized Energy Storage Capacitor

Comparing image compression predictors using fractal dimension

Design of High-linearity Delay Detection Circuit for 10-Gb/s Communication System in 65-nm CMOS

Linear PFC regulator for LED lighting with the multi-level structure and low voltage MOSFETs.

Transcription:

Inernaional Conference on Advanced Maerial Science and Environmenal Engineering (AMSEE 06) Faul Diagnosis Sysem Idenificaion Based on Impedance Maching Balance ransformer Yanjun Ren* and Xinli Deng Chongqing Radio & V Universiy, Chongqing 40005, China *Corresponding auhor ha is high volage side is conneced ino a sar connecion, he low volage side is conneced ino riangle juncion and he iron core limb winding in B phase of low volage side exend supporing arms, called balance winding. From he figure, we can see ha he core limb in A phase and C phase of impedance maching ransformer winds wo windings and ha in B phase winds four windings. In accordance o he principle of he muli-winding ransformer circui, magneic circui and equivalen circui, he linear model of impedance maching balance ransformer in equaion () can be obained hrough a series of simplified processing [7]. In equaion (), K=/N and x is he shor circui reacance of high volage side, oher noaions see he reference [8]. Absrac Impedance maching balance ransformer is ofen used as he main elecrified railway racion subsaion ransformer. In order o solve he problem ha radiional faul diagnosis mehod could no accuraely idenify he inernal or exernal failure, i is proposed an idenificaion algorihm of recursive leas square parameer esimaion online o disinguish he inernal faul from exernal faul, which is based on ha he winding resisance and shor circui reacance values have no been changed in normal operaion for impedance maching balance ransformer or exernal faul, bu hey have been changed for inernal faul. he curves of simulaion proved ha he proposed algorihm could effecively disinguish inernal faul from exernal faul. he research resuls show he feasibiliy o idenify wo sors of differen fauls. Keywords-sysem idenificaion; impedance maching balance ransformer; faul diagnosis ia ua I. INRODUCION Impedance maching balance ransformer wih ransformaion and commuaion is he main elecrified railway racion subsaion ransformer[]. ransformer s sable operaion is direcly relaed o he safey of high-speed rail, so i is very imporan o accuraely disinguish possible faul ypes and design differen proecion devices[]. ransformers of elecrified railway power supply sysem work under poor condiions. Currenly, microcompuer proecion devices in operaion are almos designed based on he principle of differenial proecion. In general, i is difficul o disinguish o be inernal failure or exernal faul, which may lead o proecion ripping or malfuncion and hen endanger he safe operaion of elecrified railway sysem[3]. Based on ha he winding resisance and shor circui reacance values could no been changed in normal operaion for impedance maching balance ransformer or exernal faul while he winding resisance and shor circui reacance values would changed dramaically, he winding resisance and shor circui reacance values of ransformer could be online idenified[4]. By comparing he idenificaion values wih he normal parameer values, we can deermine wheher he impedance maching balance ransformer proecion acion is required[5]. II. ub u ib i ib i ub ia ua N u uc ic 3 N N 3 uc N N FIGURE I. HE WINDING OF IMPEDANCE MACHING BALANCE RANSFORMER From he equaion (), we can see ha ua, ub and uc are he volage values of hree phases a he high volage side. ia, ib and ic are he curren values of windings a he high volage side. u, u and i, i are he volage values and curren values of wo phases a he low volage side respecively, which are direcly reasured as he inpu and oupu of he ransformer s parameer idenificaion model. he winding resisance R and shor circui reacance x values a he high volage side as he parameers o be idenified can consiue he grey box model of sysem idenificaion, which can respecively consruc A, B and C phases idenificaion mahemaical models of impedance maching balance ransformer hrough discreizaion. his paper only wih A- phase winding for example analyses he relevan faul diagnosis. IDENIFICAION MAHEMAICAL MODEL OF IMPEDANCE MACHING BALANCE RANSFORMER Figure I shows he winding resisance of impedance maching balance ransformer[6]. According o he operaional principle of impedance maching balance ransformer, we know 06. he auhors - Published by Alanis Press 70

3 3 ua ub Ku RiA RiB R 3 3 x dia i i K 3 3 d 5 3x 6 3 3x di B di 8 3 d 4K d 7 4 3x di 4K d K R u B u u RiB i i 3 3 K d i i 3 3K 3 3 u C u B Ku RiC Ri B R 3 3 x dic i i K 3 3 d 5 3x 6 3 3x di B di 8 3 d 4K d 7 4 3x di 4K d x dib 5 7 3 3 x d K d For online real-ime deecion he winding resisance and shor reacance values of ransformer, winding parameer idenificaion model of ransformer A-phase is as follows. z A n 3 ua n ua n ub n ub n u n u n K ia n ia n ib n ib n h A n 3 3 i n i n 3 i n i n, 3K 3K ia A 5 3 B B n i n i n i n 8 3 6 3 3 i n in 7 4 3 i n in 4K 4K A Rx, () III. FAUI DIAGNOSIS MEHOD OF IMPEDANCE MACHING BALANCE RANSFORMER BASED ON RLS A. Forgeing Facor Algorihm of Recursive leas Square Parameer Esimaion he characerisics of he leas square echnique is direcly o use all he observed daa have been obained o compue and process[9]. herefore, he parameers of sysem model need o been re-esimaed if having he new observaion daa every ime, which means ha all he obained observed daa o esimae he parameers of he model needs o be sored. And he complexiy of solving he marix equaion ˆ ( H H ) H Z resuls in he difficulies of L L L L L L online idenificaion. he soluion of his problem is o be urned ino recursive leas square algorihm. he basic idea is as follows. ( k) = ( k ) + correcion erm (3) he characerisic of he mehod is o revise he las ime esimaed resuls wih new obained observaional daa and hen recursively esimae he nex ime parameer resuls[0]. However, here will be daa sauraion phenomenon as he increase of daa when he mehod applies o online real-ime idenificaion. In oher words, as he growh of k, gain marix K (k) will gradually ends o zero so ha i could no revise ( k). Forgeing facor mehod can solve daa sauraion phenomenon effecively. he basic idea is ha he old daa adds forgeing facor o reduce he amoun of old daa informaion and increase he amoun of new daa informaion. Forgeing facor recursive leas square algorihm is as equaion (4). ˆ ˆ ( k ) k K ( k )[ z( k ) h ( k ) ˆk ] K ( k ) P k h( k ) h ( k ) P k h k P ( k ) [ I K k h ( k )] P k (4) Where is forgeing facor, 0, is no usually less han 0.9. 0.95 will be seleced if he process is linear[]. he large forgeing facor will reduce he racking performance of he algorihm, and he sysem will be easily affeced by noise wih he small forgeing facor. B. he Process of Parameer Idenificaion and Iniial Value Selecion aking A-phase as an example, he process of parameer idenificaion can be analyzed by means of idenificaion model (4). According o he measured daa, A z n, A h n, A can be obained. Subsiuing recursive leas square algorihm equaion, we can idenify he parameers. Figure II shows he program flow diagram of he algorihm. In figure II 0 is he iniial value of idenificaion parameer, idenificaion parameer value a k is he hreshold. he simulaion program of algorihm can be wrien using MALAB[]. ik is he 7

o be idenified will be analyzed in various operaing condiions. (0) P(0) za( k), h A( k) K( k), P( k), ( k) ik 0 A. Normal Operaion Figure III and figure IV shows he idenificaion curve graph of he winding resisance value R and shor circui reacance value x a he high volage side when he impedance maching balance ransformer is in normal operaion. In accordance o he parameers idenificaion curve, we can see ha when he impedance maching balance ransformer is in normal operaion, he resisance idenificaion value of he winding a he high volage side has a cerain oscillaion a 0.00s ago bu hen quickly converge near.374s. A he same ime, he shor circui reacance idenificaion value a he high volage side has a grea change a firs and hen ends o be sable. We also can see ha he winding resisance value R and shor circui reacance value x have been changed lile in normal operaion for impedance maching balance ransformer a he high volage side. FIGURE II. HE PROGRAM FLOW DIAGRAM OF HE RLS ALGORIHM In order o improve he proecion sensiiviy and rapidiy of impedance maching balance ransformer, he iniial value of he algorihm is, where is sufficienly large P 0 0 I real number, is he parameer values of impedance maching balance ransformer given by manufacurers, he forgeing facor is 0.95. IV. HE SIMULAION AND ANALYSIS he simulaion employs a impedance maching balance ransformer, whose capaciy is 5MVA, raed volage is U N U N 00 3 7.5 KV and he shor circui reacance is.7+j5.06, namely, ra.36, x 5.06. During normal operaion, he error of original and secondary side volage ransformer is %, he error of original and secondary side curren ransformer is 0% and he errors of all he resisance and reacance parameers are 0.5%. he original parameers ra, x given by he model of impedance maching balance ransformer are used as he iniial value of parameer values o be idenified in simulaion. Using elecromagneic ransien simulaion program (EMP) we can ge he real-ime volage and curren sampling values of he impedance maching balance ransformer windings on boh sides. And hen hese values are used as he inpu and oupu of he idenificaion model. he parameer values o be idenified a every sampling ime can be calculaed by wriing he program of recursive leas square algorihm in MALAB simulaion sofware. Nex he changes of impedance maching balance ransformer parameer values FIGURE III. HE IDENIFICAION CURVE FOR RESISANCE FIGURE IV. HE IDENIFICAION CURVE FOR SHOR CIRCUI REACANCE B. Exernal Phase Winding Shors o Ground Earh shor circui of exernal phase winding a he secondary side (exernal faul) suddenly occurred a 0.00s, when he impedance maching balance ransformer is in normal operaion. Under he circumsance, he idenificaion curve graph of he winding resisance value R and shor circui 7

reacance value x of impedance maching balance ransformer a he high volage side is shown as figure V and figure VI. FIGURE VII. HE IDENIFICAION CURVE FOR RESISANCE IN INERNAL faul FIGURE V. HE IDENIFICAION CURVE FOR RESISANCE IN EXERNAL faul FIGURE VIII. HE IDENIFICAION CURVE FOR SHOR CIRCUI REACANCE IN INERNAL FAUL FIGURE VI. HE IDENIFICAION CURVE FOR SHOR CIRCUI REACANCE IN EXERNAL FAUL As shown in figure 5 and figure 6, he winding resisance value R and shor circui reacance value x a he high volage side have oscillaion when Exernal phase winding shors o ground a 0.00s, bu hen he values end o be sable and almos equal o hem in normal operaion. C. A Phase Winding a he High Volage Side Shors o Ground Single phase earh faul of A-phase winding a he high volage side (inernal faul) suddenly occurred a 0.00s when he impedance maching balance ransformer is in normal operaion. Under he circumsance, he idenificaion curve graph of he winding resisance value R and shor circui reacance value x of impedance maching balance ransformer a he high volage side is shown as figure VII and figure VIII. As shown in figure VII and figure VIII, he winding resisance and shor circui reacance idenificaion values a he high volage side remain unchanged in normal operaion for impedance maching balance ransformer. he values change dramaically and will never converge when Single phase earh faul of A-phase winding a he high volage side occurred a 0.00s. he paper will no explain he iner-phase shor circui faul and iner-run shor circui faul of phase winding for impedance maching balance ransformer, however, he resuls is similar o he above menioned. he simulaion resuls show ha he winding resisance and shor circui reacance values have no been significanly changed in normal operaion for impedance maching balance ransformer or exernal faul, bu hey have been dramaically changed for inernal faul. V. CONCLUSION In order o solve he puzzle ha he exising faul diagnosis mehods could no accuraely idenify he ypes of faul for impedance maching balance ransformer, he paper proposed a faul diagnosis mehod exploiing recursive leas square algorihm o idenify he inernal failure and exernal faul based on operaional feaure of impedance maching balance ransformer. he simulaion resuls in MALAB indicae ha he proposed mehod could effecively make an accurae diagnosis of he inernal and exernal faul. REFERENCES [] IEEEGuide for Proecive Relay Applicaions o Power ransformers, 000. 73

[] K. L. Buler-Purry and A. Kuforiji "Experimenal resuls from shor-circui fauls ona disribuion ransformer", Proc. IEEEransm. Disrib. Conf. Expo., vol., pp.99-306 00. [3] K. L. Buler-Purry, M. Bagriyanik, M.J. Mousavi and P. Palmer-Buckle "Experimenal invesigaion of inernalshor circui fauls leading o advanced incipien behavior and failureof a disribuion ransformer", Proc.IEEE Power Sys. Conf. Expo., vol. 3, pp.407-46 004. [4] V. Behja and A. Vahedi "An experimenal approachfor invesigaing low-level inerurn winding fauls in power ransformers", Elec. Eng., vol. 95, no., pp.35-45 03. [5] R. S. Bhide, M.S. S. Srinivas, A. Banerjee and R. Somakumar "Analysis of windinginer-urn faul in ransformer: A review and ransformer models", Proc. IEEE In. Conf. Susain. Energy echnol., pp. -7 00. [6] N. C. Joshi, Y.R. Sood, R.K. Jarial and R. hapliyal "ransformer inernal winding faulsdiagnosis mehods: A review", MI In.J. Elec. Insrum. Eng., vol., no., pp.77-8 0. [7] A. Bhoomaiah, P. K. Reddy, K. S. L. Murhy, P. A. Naidu and B. Singh "Measuremen of neuralcurrens in a power ransformer and faul deecion using waveleechniques", Proc. Annu. Rep. Conf.Elec. Insul. Dielecr. Phenomena, pp.70-73 004. [8] M. Faridi, E. Rahimpour, M. Kharezi, H. R. Mirzaei and A. Akbari "Localizaion of urn-o-urnfaul in ransformers using arificial neural neworks and windingransfer funcion", Proc. IEEE 0hIn. Conf. Solid Dielecr., pp. -4 00. [9] V. Behja and A. Vahedi "Numerical modeling of ransformers inerurn faulsand characerizing he fauly ransformer behavior under various faulsand operaing condiions", IE Elec.Power Appl., vol. 5, no. 5, pp.45-43 0. [0] M. Vikash, K. Sharlinprija and M.K. Ilampoornan "ransformer faul deecion by frequencyresponse analysis", In. J. Compu.Sci. Elec. Eng., vol., no., 0. [] M. F. Cabanas, F.P. Gonzales, M. G. Melero, C. H. Rojas, G. A. Orcajo, J. M. Cano and J. G. Norniella "Insulaion faul diagnosisin high volage power ransformers by means of leakage flux analysis", Progr. Elecromagn. Res., vol. 4, pp. -34 0. [] L. M.R. Oliveira and A. J. M. Cardoso "A permeance based ransformer modeland is applicaion o winding inerurn arcing faul sudies", IEEE rans. Power Del., vol. 5, no. 3, pp.589-598 00. 74