变压器油中气体拉曼光谱检测及信号处理方法

准确检测变压器油中溶解故障特征气体是诊断变压器运行状态的重要技术手段之一。论文基于拉曼光谱和腔长调制频率锁定原理,搭建了变压器故障特征气体频率锁定腔增强拉曼光检测平台,实现了H_2、CH_4、C_2H_2、C_2H_4、C_2H_6、CO、CO_2等七种故障特征气体的同时检测;1atm时,H_2、CH_4、C_2H_2、C_2H_4、C_2H_6、CO、CO_2的最小检测浓度实验值分别达到106、25、45、73、41、170、126(ppm)。频率锁定增强腔技术使最小检测浓度提高了约68倍。运用小波模极大值法对H_2的拉曼光谱检测信号进行了去噪处理,提出了基于包络线迭代法的光谱基线校正方法,校正后的光谱荧光背景残留减少,使气体拉曼光谱检测准确度提高了约2.95%,为变压器油中溶解故障气体同时准确检测提出了一种新方法。

Raman spectroscopy detection and signal processing method for the gases in transformer oil

Accurate detecting the dissolved fault characteristic gases in transformer oil is one of the important technical means to diagnose transformer operating condition. On the basis of Raman spectroscopy detection technique and cavity length modulation frequency locking principle, in this paper a frequency locking cavity enhanced Raman spectroscopy detection platform for transformer fault characteristic gas is proposed, which can simultaneously detect seven fault gases i.e. H2, CH4, C2H2, C2H4, C2H6, CO and CO2 in transformer oil. The experiment values of the minimum detectable concentration of H2, CH4, C2H2, C2H4, C2H6, CO and CO2 reach approximately 106, 25, 45, 73, 41, 170 and 126 ppm,respectively. The frequency locking cavity enhanced technology improves the minimum detectable concentration by about 68 times. The wavelet modulus maxima method is used to denoise the Raman spectrum detection signal of H2, and the spectrum baseline correction method based on the envelope line iteration method is proposed. After correction with the envelope line iteration method, the residue of the spectrum fluorescence background is reduced, and the gas Raman spectrum detection accuracy is improved by about 2.95%. The proposed method provides a new way for accurately detecting the fault gases dissolved in transformer oil simultaneously.