Font Size: a A A

Transformer Mechanical Vibration Detection Based On Optical Frequency Domain Reflectometry

Posted on:2022-08-23Degree:MasterType:Thesis
Country:ChinaCandidate:M ZhangFull Text:PDF
GTID:2492306338961499Subject:Electrical engineering
Abstract/Summary:PDF Full Text Request
Vibration signal analysis method is a highly sensitive on-line detection method of transformer mechanical state.Traditional acceleration sensors have shortcomings such as low anti-electromagnetic interference ability,poor multiplexing ability,and inability to distributed sensing.In order to solve the above problems,this thesis proposes a distributed transformer vibration sensing method based on Optical Frequency Domain Reflectometry(OFDR).First of all,this thesis establishes a mathematical model of the principle of optical frequency domain reflectometry,and masters the limiting factors of OFDR detection frequency.The traditional OFDR is limited by the repetition frequency of the tunable light source sweep,and cannot detect the vibration signal of the transformer.In order to solve the above problems,this thesis establishes a new vibration detection system that integrates OFDR and fiber interference and studies the phase modulation principle of vibration signal to optical signal.An OFDR distributed vibration signal demodulation algorithm based on differential cross multiplication is proposed.The OFDR sensing system established in this thesis transforms the traditional OFDRs light intensity measurement into an optical phase measurement,breaking through the limitation of tunable laser source parameters,and solving the problem that OFDR cannot detect high-frequency vibration signals.Then,this thesis analyzes the factors affecting the spatial resolution and upper limit of the detection frequency of the proposed OFDR distributed vibration sensor.The frequency sweep rate of the laser source and the interval between adjacent reflectors are optimized.This thesis determines the filter bandwidth range of the demodulation algorithm.This thesis carries out simulation analysis and research,using single-frequency sinusoidal signal and measured vibration signal as input sources to carry out theoretical exploration on the performance of the proposed OFDR distributed vibration sensing system.The simulation results show that in the case of two vibration sources,the correlation coefficients of the demodulated signal and the original signal are all above 0.9;in the case of 10 vibration sources,the correlation coefficients of the 10 demodulated vibration signals and the original vibration signal are all above 0.8.Realize the accurate positioning of the vibration position.Finally,this thesis develops an optical fiber vibration sensor,builds an OFDR vibration sensor detection platform,and develops an OFDR vibration detection system with freely set spatial resolution and not restricted by system parameters.The sensor calibration experiment results show that the developed optical fiber vibration sensor has a sensitivity of 0.2405 rad/g,a minimum detection limit of 0.01g,and a response fluctuation of about 5 dB within a 1 kHz bandwidth.The downward trend of the frequency response curve is corrected by an interpolation restoration algorithm.Multi-source vibration signals sensing experiment proves that the system can accurately detect and locate multiple vibration signals at the same time.The vibration measurement experiment of the transformer in field operation is carried out.The experiment results show that the frequency components of the detection signal of the optical fiber vibration sensor and the standard acceleration sensor are the same,and the maximum error of the acceleration amplitude of each frequency component does not exceed 10%,which proves that the system meets the needs of actual transformer operation vibration detection.Compared with the traditional electronic system,this system has the advantages of strong anti-electromagnetic interference ability and distributed sensing.
Keywords/Search Tags:transformer, vibration sensing, optical frequency domain reflection, distributed
PDF Full Text Request
Related items