Font Size: a A A

Analysis Of Vibration State Of Transformer And Fiber Bragg Grating Monitoring

Posted on:2017-05-08Degree:MasterType:Thesis
Country:ChinaCandidate:K LiFull Text:PDF
GTID:2132330488465648Subject:Measuring and Testing Technology and Instruments
Abstract/Summary:PDF Full Text Request
As a key part of power industry, power transformer is the key part of voltage conversion function in the normal operation of power industry. The Core and winding are the main factors that generated the transformer vibration, so vibration signal contains a large number of core and winding condition information, and it can be predicted and diagnosed the running state of the transformer through collecting vibration signal of transformer. At the same time, the corrosion resistance and other characteristics of fiber grating can be used for layouting of sensors to the transformer interior so that the signal with a greater signal to noise ratio can be obtained.The main research contents of this paper contents:1. First, the basic working principle of the transformer is analyzed, also the core and winding vibration mechanism and vibration cause is analyzed. The vibration of the core is mainly caused by magnetostrictive phenomenon and the main reason for the winding vibration is caused by the electric field force, also the vibration is 2 times the power of the fundamental frequency which is 100Hz frequency and is mainly distributed in the range of 100 Hz to 500 Hz. After that, the vibration signal into optical fiber wavelength shift is analyzed, and fiber Bragg grating wavelength change and acceleration is linear relationship is obtained according to the concrete structure analysis. Finally, the selected sensor sensitivity is 82pm/g and working frequency is 50 to 550 Hz is obtained through the experiment, so it can basically meet the requirement of monitoring of transformer.2. Based on the analysis of previous work, the core model and core-winding model is established by ANSYS software according to the actual parameters of the transformer and respectively constraints is applied according to the core and core-winding is fixed on the ground actually, also calculation of core and core-winding model is finished through the finite element analysis. The core model of the top three natural frequencies are 131.1 lHz, 280.12Hz,399.27Hz, and the core-winding model top three natural frequencies are 70.807Hz,142.59Hz,163.73Hz, also the corresponding mode is got. Through the analysis of vibration modes obtained, the sensors should be laid on the top to the core so that greater signal can be obtained. At the same time, through the analysis of the harmonic response signal, people should pay attention to the two directions, to provide a theoretical basis for further practical monitoring in the layout process.3. In the test of monitoring,35kV, S13-12500/35 type oil immersed transformer is selected. The fiber Bragg grating vibration sensors arranged inside the transformer is fixed on the top of the transformer core respectively, and the sensitive direction of the transformer in two directions should correspond to the sensors. The External, of the sensors is connected to the fiber Bragg grating high speed demodulator and PC to collect and monitor the signal. First the no-load operation of vibration signal is measured, after that, the load operation of vibration signal is measured in the state of the load monitoring. Two groups of load signal are separated through the JADE algorithm for the spectrum characteristics analysis. We find the errors of the decomposition vibration signal 1 which is from 100Hz to the harmonic components relative to the no-load vibration data are 0%, 18%,9%,54% and 17%, and the average relative errors is 19.6%. Also the third natural frequency of the core is 399.27Hz is found, so it may be the factor which caused vibration signal of the core in the vicinity of 400Hz has the large amplitude.
Keywords/Search Tags:Transformer vibration, Condition monitoring, Fiber grating, Finite element analysis, Point selection, Blind source separation, JADE algorithm
PDF Full Text Request
Related items