Gas Detection Based On TDLAS-WMS And Its Application In Oil Immersed Transformer Fault Detection

The transformer is one of the most important substation equipment in the power system.The normal operation of the transformer guarantees the overall safety and economy of the power system.Therefore,it is especially important to find transformer faults in time.Insulation oil is used in most of the current transformers.Its function is heat dissipation,insulation and arc suppression.When the internal fault occurs in the transformer,some characteristic gases are generated in the oil.Firstly,according to previous research,the application status and advantages of dissolved gas detection technology and TDLAS technology in transformer oil are discussed and analyzed.The principle of dissolved gas generation in transformer oil,the gas production and dissolution process in oil were discussed in depth.The corresponding relationship between the type of dissolved characteristic gas in oil and the type of transformer failure was clarified.Secondly,according to the theoretical model of TDLAS-WMS,aiming at the cumbersome problem of traditional lock-in amplification harmonic extraction method,a harmonic demodulation technique based on spectrum analysis method is proposed.After the theoretical derivation of the method,the numerical simulation is carried out,and the simulation results are consistent with the theory,which proves the accuracy of the proposed method.In order to further prove the accuracy and effectiveness of the proposed method,the TDLAS experimental device was built and the experiment corresponding to the numerical simulation was carried out.The experimental results show that the proposed spectrum analysis method has the same harmonic demodulation accuracy as the traditional lock-in amplification method.And when the methane gas concentration is lower than 30 ppm,the spectrum analysis method has better anti-noise performance and higher harmonic demodulation precision.Thirdly,in order to eliminate the influence of the initial intensity fluctuation on the TDAS trace gas detection system,the combination method of spectrum analysis method and the harmonic ratio method is proposed which based on the spectrum analysis method.The feasibility of the method is theoretically deduced.Theoretically,the first harmonic is independent of the gas concentration at the center frequency,and the second harmonic is proportional to the gas concentration at the center frequency,so the ratio of the two eliminates the effect of fluctuations in light intensity and is maintained in direct proportion to the concentration of the gas to be measured.The accuracy and effectiveness of the demodulation method are also proved from the aspects of simulation and experiment.The method eliminates the influence of light intensity fluctuation and can be used in a practical environment with high noise.Finally,Aiming at the problem of transformer fault detection by detecting the characteristic gas released from oil,a trace methane gas detection device in transformer oil based on TDLAS technology was designed and built.The two data processing methods proposed are applied to process the transmitted light intensity signals during normal and fault conditions.Comparing the experimental results,it is proved that the difference between the measurement results obtained by the two data processing methods under laboratory conditions is less than 2 ppm,so it is simple to apply the first processing method under stable conditions.When the actual measurement environment noise has a large influence,the second method is applied to make the detection result more accurate.At the same time,it proves that the technology has good feasibility in transformer fault detection,and provides preliminary research and assistance for the establishment of transformer fault feature knowledge base and fault detection development.