Research on fault prediction method of power electronic circuits based on least squares support vector machine

Similar documents
Deterioration analysis of aluminum electrolytic capacitor for DC-DC converter

Four-dimensional hyperchaotic system and application research in signal encryption

Research Article Time Series Adaptive Online Prediction Method Combined with Modified LS-SVR and AGO

Estimation of Circuit Component Values in Buck Converter using Efficiency Curve

Vibration and Modal Analysis of Small Induction Motor Yan LI 1, a, Jianmin DU 1, b, Jiakuan XIA 1

Multi-wind Field Output Power Prediction Method based on Energy Internet and DBPSO-LSSVM

Nonlinear Controller Design of the Inverted Pendulum System based on Extended State Observer Limin Du, Fucheng Cao

ARX System identification of a positioning stage based on theoretical modeling and an ARX model WANG Jing-shu 1 GUO Jie 2 ZHU Chang-an 2

Nonlinear dynamic simulation model of switched reluctance linear machine

NON-FIXED AND ASYMMETRICAL MARGIN APPROACH TO STOCK MARKET PREDICTION USING SUPPORT VECTOR REGRESSION. Haiqin Yang, Irwin King and Laiwan Chan

The Fault extent recognition method of rolling bearing based on orthogonal matching pursuit and Lempel-Ziv complexity

Converter System Modeling via MATLAB/Simulink

Research Article Stabilization Analysis and Synthesis of Discrete-Time Descriptor Markov Jump Systems with Partially Unknown Transition Probabilities

Electrolytic Capacitor Age Estimation Using PRBS-based Techniques

Analysis on the Operating Characteristic of UHVDC New Hierarchical Connection Mode to AC System

Anti-synchronization of a new hyperchaotic system via small-gain theorem

Research on Permanent Magnet Linear Synchronous Motor Control System Simulation *

Support Vector Machine Regression for Volatile Stock Market Prediction

Open Access Permanent Magnet Synchronous Motor Vector Control Based on Weighted Integral Gain of Sliding Mode Variable Structure

Influence Analysis of Transmission Lines Insulator on the Conductor Ice-shedding

Discussion About Nonlinear Time Series Prediction Using Least Squares Support Vector Machine

Experimental and numerical simulation studies of the squeezing dynamics of the UBVT system with a hole-plug device

A prognosis case study for electrolytic capacitor degradation in DC-DC converters

International Journal of Advance Engineering and Research Development SIMULATION OF FIELD ORIENTED CONTROL OF PERMANENT MAGNET SYNCHRONOUS MOTOR

THE power transfer capability is one of the most fundamental

Bearing fault diagnosis based on TEO and SVM

Investigation of Field Regulation Performance of a New Hybrid Excitation Synchronous Machine with Dual-Direction Magnetic Shunt Rotor

Fault prediction of power system distribution equipment based on support vector machine

Experimental Studies of Ageing in Electrolytic Capacitors

A Wavelet Neural Network Forecasting Model Based On ARIMA

Modeling, Analysis and Control of an Isolated Boost Converter for System Level Studies

Research of Hybrid Three-phase equilibrium Technology

Forecast daily indices of solar activity, F10.7, using support vector regression method

Study of Chaos and Dynamics of DC-DC Converters BY SAI RAKSHIT VINNAKOTA ANUROOP KAKKIRALA VIVEK PRAYAKARAO

Two-stage Pedestrian Detection Based on Multiple Features and Machine Learning

1439. Numerical simulation of the magnetic field and electromagnetic vibration analysis of the AC permanent-magnet synchronous motor

Outliers Treatment in Support Vector Regression for Financial Time Series Prediction

A Hybrid Time-delay Prediction Method for Networked Control System

Analysis of Electrolytic Capacitor Degradation under Electrical Overstress for Prognostic Studies

An Improved Quantum Evolutionary Algorithm with 2-Crossovers

Discussion of Some Problems About Nonlinear Time Series Prediction Using ν-support Vector Machine

Dynamic Data Modeling of SCR De-NOx System Based on NARX Neural Network Wen-jie ZHAO * and Kai ZHANG

Generator Thermal Sensitivity Analysis with Support Vector Regression

EMC Considerations for DC Power Design

Space-Time Kernels. Dr. Jiaqiu Wang, Dr. Tao Cheng James Haworth University College London

Storage Lifetime Prognosis of an Intermediate Frequency (IF) Amplifier Based On Physics of Failure Method

Speed Control of PMSM Drives by Using Neural Network Controller

A New Robust Decentralized Control Method for Interconnected Nonlinear Systems Based on State Extension and Adaptive Tracking

Research Article Data-Driven Fault Diagnosis Method for Power Transformers Using Modified Kriging Model

SYNCHRONIZATION CRITERION OF CHAOTIC PERMANENT MAGNET SYNCHRONOUS MOTOR VIA OUTPUT FEEDBACK AND ITS SIMULATION

An Improved Method of Power System Short Term Load Forecasting Based on Neural Network

Wind Turbine Gearbox Temperature Monitor Based On CITDMFPA-WNN Baoyi Wanga, Wuchao Liub, Shaomin Zhangc

1863. Active-disturbance rejection control based on a novel sliding mode observer for PMSM speed and rotor position

2183. Vector matching-based disturbance rejection method for load simulator

On the average diameter of directed loop networks

Research Article Design of PDC Controllers by Matrix Reversibility for Synchronization of Yin and Yang Chaotic Takagi-Sugeno Fuzzy Henon Maps

Bearing fault diagnosis based on Shannon entropy and wavelet package decomposition

Power Electronic Circuits Design: A Particle Swarm Optimization Approach *

The Phase Detection Algorithm of Weak Signals Based on Coupled Chaotic-oscillators Sun Wen Jun Rui Guo Sheng, Zhang Yang

Information. A Fitting Approach. 1. Introduction

No. 6 Determining the input dimension of a To model a nonlinear time series with the widely used feed-forward neural network means to fit the a

Electronic Supplementary Information

Advanced Technology of Electrical Engineering and Energy EMI PCB

Yang Yu, Xudong Guo, and Zengqiang Mi. 1. Introduction

Adaptive Perturb & Observe MPPT Algorithm of PV Array Based on Hysteresis Comparison

LECTURE 8 Fundamental Models of Pulse-Width Modulated DC-DC Converters: f(d)

bifunctional electrocatalyst for overall water splitting

Supporting Information

PoS(CENet2017)018. Privacy Preserving SVM with Different Kernel Functions for Multi-Classification Datasets. Speaker 2

Journal of Engineering Science and Technology Review 7 (1) (2014)

Laser on-line Thickness Measurement Technology Based on Judgment and Wavelet De-noising

Transient Thermal Analysis of MCM towards Understanding Failure Mechanism of Intermittent Faults

Modeling and sliding mode control of electric vehicle charger control system

The output voltage is given by,


Logging characteristic analysis of basalt in eastern depression of Liaohe Oilfield

2262. Remaining life prediction of rolling bearing based on PCA and improved logistic regression model

DISTURBANCE LOAD MODELLING WITH EQUIVALENT VOLTAGE SOURCE METHOD IN GRID HARMONIC ASSESSMENT

Curriculum Vitae Bin Liu

Chapter 3. Steady-State Equivalent Circuit Modeling, Losses, and Efficiency

Wind Speed Forecasting in China: A Review

An Efficient Bottom-Up Extraction Approach to Build the Behavioral Model of Switched-Capacitor. ΔΣ Modulator. Electronic Design Automation Laboratory

High Performance and Reliable Torque Control of Permanent Magnet Synchronous Motors in Electric Vehicle Applications

Computationally efficient models for simulation of non-ideal DC DC converters operating in continuous and discontinuous conduction modes

Application Research of ARIMA Model in Rainfall Prediction in Central Henan Province

Condition Based Maintenance Optimization Considering Improving Prediction Accuracy

Mathematical Modeling and Dynamic Simulation of a Class of Drive Systems with Permanent Magnet Synchronous Motors

Bayesian Remaining Useful Lifetime Prediction of Thermally Aged Power MOSFETs

Weighted Fuzzy Time Series Model for Load Forecasting

Open Access Variable Step Length Incremental Conductance MPPT Control Based on the Power Prediction

Robust Speed Controller Design for Permanent Magnet Synchronous Motor Drives Based on Sliding Mode Control

Homework Assignment 11

Modelling and Teaching of Magnetic Circuits

ANALYSIS OF SMALL-SIGNAL MODEL OF A PWM DC-DC BUCK-BOOST CONVERTER IN CCM.

MICROGRID is a future trend of integrating renewable

Supporting Information. Graphene Textile Strain Sensor with Negative Resistance Variation for Human Motion

Implementation Possibilities of SMD Capacitors for High Power Applications

Research Article Finite Element Analysis of Flat Spiral Spring on Mechanical Elastic Energy Storage Technology

EE105 Fall 2014 Microelectronic Devices and Circuits

Transcription:

15 8 2011 8 ELECTRI C MACHINES AND CONTROL Vol. 15 No. 8 Aug. 2011 LS-SVM 1 2 1 1 1 1. 210016 2. 232001 least squares support vector machine LS-SVM Buck LS-SVM LS-SVM 2% TP 206 A 1007-449X 2011 08-0064- 05 Research on fault prediction method of power electronic circuits based on least squares support vector machine JIANG Yuan-yuan 1 2 WANG You-ren 1 CUI Jiang 1 SUN Feng-yan 1 1. College of Automation Engineering Nanjing University of Aeronautics and Astronautics Nanjing 210016 China 2. College of Electric and Information Engineering Anhui University of Science and Technology Huainan 232001 China Abstract Aiming at the issue of fault prediction technique of power electronic circuits a method based on characteristic parameter data and least squares support vector machine LS-SVM for the prediction of power electronic circuits was proposed. Taking the Buck converter circuit as an example the fault prediction of power electronic circuits was achieved. Firstly the output voltage was selected as monitoring signal and then the average voltage and ripple voltage were extracted as characteristic parameters. Lastly LS-SVM algorithm was used to predict Buck converter circuit. The experimental results show that the LS- SVM algorithm is especially accurate in predicting the average voltage and ripple voltage with the relative error less than 2%. The new method can trace the characteristic parameters trend and can be effectively applied in fault prediction of power electronic circuits. Key words power electronic circuits fault prediction characteristic parameter data driving least squares support vector machine LS-SVM 2010-07 - 24 60871009 2009ZD52045 CXLX11-0183 NS2010063 1982 1963 1977 1983

8 LS-SVM 65 0 1-2 3 1 Buck Fig. 1 Buck converter circuit 1 Buck DC-DC 4 5 6 3 Buck Buck 2 a u o 7-9 1 U o N u o i i U o = 1 u least squares support N N o i 1 i = 1 vector machine LS-SVM b - LS-SVM - Δu 1 1. 1 LS - SVM SVM 10 11 support vector machine regression SVR LS-SVM 1 x Buck Hilbert 1 MOSFET IRF151 SVM f = 50 khz D = 0. 22 L = 43 μh f x = w T x + b w R n b R 2 D1 MUR405 C = 220 μf ESR R L = 1. 25 Ω 1. 2 LS-SVM f x

66 15 x w w T w b R n n R support vector SV 7 2 minj w e = 1 w b e 2 w 2 + γ n e 2 i } i = 1 3 1 Buck S. t. y i = w T x i + b + e i Pspice e i γ x i x = u o x 1 x i x n i y i y = Matlab7. 6 y 1 y i y n i Lagrange Lagrange α i L w e b α = J w e - n i = 1 α i w T x i + b + e i - y i 4 Karush-Kuhn-Tucker LS- SVM f x = n i = 1 α i K x i x + b 5 K x y = x y Mercer Sigmoid 1. 3 LS-SVM 2 2 Fig. 2 Flowchart of the fault prediction 1 2. 1 Buck 1 h 5 h 24 h 2 8 12 C 6 LS-SVM 2 LS-SVMlab Buck ESR L MOSFET R ds 3 V th g m 1 1 4 1 Buck 5 LS-SVM 2

8 LS-SVM 67 Table 1 1 Buck LS-SVM The parameters of the components in Buck Buck /h C /μf R ESR /mω L /μh R ds /mω V th /V g m /μ 1 220 512 43 1. 6 2. 83 20. 5 2 218 523 42. 9 3. 6 2. 85 20. 4 3 215 532 42. 8 5. 7 2. 87 20. 3 4 212 544 42. 6 8. 0 2. 89 20. 2 5 208 557 42. 4 11 2. 91 20. 1 6 205 570 42. 2 15 2. 93 20 7 200 583 42 19 2. 95 19. 9 8 194 608 41. 7 24 2. 97 19. 7 9 189 621 41. 5 29 3. 0 19. 5 10 183 637 41. 1 34 3. 03 19. 2 Table 3 3 Prediction results of characteristic parameters Table 2 2 The characteristic parameters data /h /V /V 1 5. 137 0. 768 2 5. 135 0. 782 3 5. 132 0. 793 4 5. 130 0. 809 5 5. 126 0. 825 6 5. 123 0. 842 7 5. 118 0. 858 8 5. 112 0. 889 9 5. 106 0. 905 10 5. 100 0. 928 Fig. 3 3 Prediction curves of average voltage 2. 2 LS-SVM 1 2 1 ~ 6 7 7 1 2 ~ 7 6 8 4 LS-SVM 2 Buck gam = Buck 1003 3 LS-SVM Buck 2% U o Δu 3 4 + 1 ~ 6 7 ~ 10 4 3 4 5. 137 8 V Fig. 4 4 Predicted curves of ripple voltage 0. 5 V 10% 1 V

68 15 0. 7 V 3 4 0. 5 V 1 V 4 h 3 SVM SVM SVM March 7-14 1 gam 2 gam sig2 RBF tana USA. 2005 3585-3591. 4 4 Table 4 4 Prediction effect of different kernel function % % gam = 10 0. 065 7 0. 941 9 gam = 100 0. 048 9 0. 819 6 gam = 100 sig2 = 10 gam = 1000 sig2 = 50 0. 059 4 1. 218 9 0. 029 8 0. 543 4 and Decision 2006 21 1 77-80. 4 3 1 LALL P HANDE M BHAT C et al. Prognostics health monitoring PHM for prior-damage assessment in electronics equipment under thermo-mechanical loads C / /Proceedings of Electronic ference and Exposition February 25 - March1 2007 LS- USA. IEEE 2007 1057-1061. SVM Buck mode power converters J tronics 2008 55 1 400-406. 10 62-66. Components and Technology Conference May 29 - June1 2007 New York USA. 2007 1097-1111. 2 MA Zhangshan. A new life system approach to the prognostic and health management PHM with survival analysis dynamic hybrid fault models evolutionary game theory and three-layer survivability analysis C / /Proceedings of IEEE Aerospace Conference 2009 Big Sky Montana USA. 2009 1-20. 3 ORSAGH R BROWN D ROEMER M. Prognostic health management for avionics system power supplies C / /Proceedings of IEEE Aerospace Conference March 5-12 2005 Big Sky Mon-. AR J. 2009 30 1 91-95. LV Kehong QIU Jing LIU Guanjun. Research on life prognosis method for electronics based on dynamic damage and optimization AR model J. Acta Armamentarii 2009 30 1 91-95. 5. BP J. 2009 26 1 52-54. ZOU Xinyao YAO Ruohe. Life prediction of electronic devices based on forecast system of back propagation neural network J. Microelectronics and Computer 2009 26 1 52-54. 6. LS-SVM J. 2006 21 1 77-80. JIANG Tianhan ShU Jiong. Multi-step prediction of chaotic time series using the least squares support vector machine J. Control 7 AMARAL A CARDOSO A. Use of ESR to predict failure of output filtering capacitors in boost converters C / /Proceedings of IEEE International Symposium on Industrial Electronics May 4-7 2004 New York USA. IEEE 2004 1309-1314. 8 IMAM A DIVAN D HABETLER T et al. Real-time condition monitoring of the electrolytic capacitors for power electronics applications C / /Proceedings of IEEE Applied Power Electronics Con- New York 9 CHEN Yaow-Ming CHOU Hsu-Wei LEE Kungyen. Online failure prediction of the electrolytic capacitor for LC filter of switching-. IEEE Transactions on Industrial Elec-. J. 2007 27 10 74

74 15 11 LIU Jun HUANG Mengzhi WANG Yang. Research on vectorcontrol system of PMSM based on internal model control of current loop C / /Second International Workshop on Computer Science and Engineering Octorber 28-30 2009 Qingdao China. IEEE 2009 297-301. 12 LENNART Harnefors HANS Peter Nee SHARMA C. Modelbased current control of AC machines using the internal model control method J. IEEE Transactions on Industry Applications 1998 34 1 133-140. 13. J. 2010 14 1 61-65. ZHU Xirong ZHOU Yuanshen FU Xiao. Three-degree-freedom internal model dynamic decoupling control of synchronous motor J. Electric Machines and Control 2010 14 1 61-65. 14. H J. 2000 29 6 526-531. LI Hua HOU Yansong. Design of internal model control feedback filter by H optimization J. Electric Machines and Control 2000 29 6 526-531. 檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪 68 CUI Jiang WANG Youren LIU Quan. The technique of power electronic circuit fault diagnosis based on higher-order spectrum analysis and support vector machines J. Proceedings of the Chinese Society for Electrical Engineering 2007 27 10 62-66. 11 MULLER K SMOLA A RATSCH G et al. Predicting time series with support vector machines C / /Proceedings of International Conference on Artificial Neural Networks October 8-10 1997 Berlin Germany. Springer 1997 999-1004. 12. MOSFET J. 2008 42 12 49-51. WANG Cailin SUN Cheng. Analysis of high characteristics and SOA of power MOSFET J. Power Electronics 2008 42 12 49-51.