Dissertation > Excellent graduate degree dissertation topics show

Partial Discharge Detection Method Using Optical-fiber Current Sensor

Author: LuYuHang
Tutor: DuBoXue
School: Tianjin University
Course: High Voltage and Insulation Technology
Keywords: PD Faraday magneto-optical effect Fiber-optic current sensors Electromagnetic interference Time-Frequency Analysis Fractal dimension Pattern Recognition
CLC: TM835
Type: PhD thesis
Year: 2008
Downloads: 608
Quote: 7
Read: Download Dissertation

Abstract


Partial discharge detection can be promptly reflected power equipment insulation aging, condition monitoring and fault diagnosis is an important means. However, due to the strong electromagnetic field interference seriously affect the sensor sensitivity and detection reliability, high performance sensors developed PD line detection is always an important topic in the study. Optical detection methods interference ability, good insulation properties, high sensitivity for detection of PD Online Research is significant. This paper will be based on principles developed by Faraday magneto-optical effect used in fiber-optic current sensor PD detection to detect the amount of light signals for partial discharge optical detection method laid the foundation. Analysis of the different types of fiber-optic current sensor characteristics, the use of Jones matrix of the fiber current sensor polarizing system research, given a different polarization states of the normalized Jones vector of the optical device of this derivation of the corresponding Jones matrix, and ultimately establish a mathematical model of fiber-optic current sensors. Development of an effective bandwidth of 300 MHz all-fiber optical fiber current sensor, the design of the signal processing circuit; double solenoid structure to improve sensitivity of the sensor conductor and the energized solenoid shape relative position between the optical path changes in the relationship does not affect the test results; analysis of its frequency response characteristics, immunity characteristics and loss characteristics of the optical path, compared with the traditional fiber optic current sensor, the detection system frequency response, high sensitivity. When the sensor structure using all-fiber optical design closure and satisfy Ampere's law, its resistance to electromagnetic interference optimal performance. And the introduction of linear birefringence for a variety of reasons; analyzes the system of linear birefringence, that the linear birefringence in the fiber affect the current sensor measurement error factors, along with the linear birefringence increases, the system sensitivity decreases, if the system is not linear birefringence, the other sources of error will be easy to predict and handle, and discusses possible measures to eliminate linear birefringence. Established a partial discharge test system and tip boards discharge, discharge plate board, internal discharge, surface discharge and suspended discharge five typical PD model developed through this fiber-optic current sensors collect a variety of partial discharge signals, experimental results show that in the strong interference environment compared with other methods, based on fiber-optic current sensor partial discharge detection with anti-interference ability, good insulation properties and fast response characteristics. Feature extraction parameters are properly carried out partial discharge pattern recognition of key aspects of the fiber-optic current sensor to the partial discharge time-domain signal into a three-dimensional spectrogram, a comprehensive characterization of partial discharge pulse time component, the frequency components and the discharge energy The distribution of the extracted characteristic parameters reflect both the time and frequency characteristics of the original signal, interference between each component, so more accurate. In this paper, based on nonlinear theory of fractal box dimension and vacancy rates for describing complex time-frequency surface, when extracted partial discharge dimensional time-frequency spectrum characteristic parameters to describe the peak of the spectrum when the steepness of the change and falling tendency to simplify the dimension of the feature vector. Time-frequency analysis method based on fractal theory and BP neural network can effectively identify different types of partial discharge, fiber-optic current sensors can be used to prove partial discharge detection and fault diagnosis.

Related Dissertations

  1. Research on Text Classification Based on Biomimetic Pattern Recongnition,TP391.1
  2. Multi-walled Carbon Nanotubes Au _AT_ of Pt, Au _AT_ of Pd core-shell structure of the catalyst preparation and electrochemical properties of,O643.36
  3. The Study of Anode Catalyst Pd/C for Direct Formic Acid Fuel Cell,O643.36
  4. Pd-based cathode catalysts of direct methanol fuel cell,O643.36
  5. Research on Identification System of Cashmere and Wool Fiber,TS101.921
  6. Intrusion detection based on the ultrasonic echo envelope in the military security patrols,E919
  7. Mouse anti-human PD-L1 functionality Xing monoclonal antibody developed by and its biological characteristics of research,R392
  8. Mouse anti-human PD-L2 monoclonal antibody and the human soluble PD-L2ELISA kit for,R392
  9. Expression and Clinic Significance of PD-1/PD-L1 Costimulatory Molecules in Peripheral Blood of Rheumatoid Arthritis,R593.22
  10. Algorithm Research for Robot Trajectory Tracking Control,TP242
  11. Expression of PD-1 on T Lymphocytes and PD-L1 on Peripheral Blood Mononuclear Cells from the Patients with Systemic Lupus Erythematosus,R593.241
  12. Inhibitory Effect of α-lipoic Acid on Inhibiting the PD-L1 in Human Umbilical Vein Endothelial Cells under Advanced Glycosylation End Products,R587.2
  13. Expression of PD-1/PD-L1 on the Lymphocytes from Peripheral Blood and Its Clinical Significance in Type 2 Diabetes Disease,R587.1
  14. Research on Key Technologies of Turning Expert System Based on SE300,TP182
  15. Improved Mathematical Model of Spatial Information Processing Based on Fuzzy Technology,O159
  16. Catalytic Study of Hydrogen Peroxide Synthesis Directly from H2 and O2,TQ123.6
  17. The Study of B7-H1/PD-1 Inhibitory Pathway and Relationship to Regulatory T Cells in Human Colon Carcinoma,R392
  18. The EMG Pattern Recognition System Based on Generalized Dynamic Fuzzy Neural Network,TP183
  19. Design and Implementation of System of Tables with Handwritten Content,TP391.41
  20. The Non-Linear Mixed Effect Models in Population Pharmacokinetics and Its Discussion Based on Stochastic Differential Equations,R911
  21. Action surface EMG nonlinear characteristic of,TH772

CLC: > Industrial Technology > Electrotechnical > High Voltage Engineering > High voltage test equipment and measurement techniques > High - voltage measurement technique
© 2012 www.DissertationTopic.Net  Mobile