Research On Dissolved Characteristic Components Sensing In Transformer Oil Based On Micro-nano Fiber Evanescent Field

Detection of dissolved characteristic components in transformer oil is helpful for sensing and evaluating the operating conditions and insulation status of oil-immersed power transformer,which has engineering value for the condition inspection and operation maintenance of transformer.Based on the principle of MNF evanescent field,the detections of dissolved characteristic components in transformer oil were studied.MNF was used as sensor to detect the moisure and acetylene in transformer oil.The experiment results proved the feasibility and the detection results were great,which had great practical significance.Firstly,the theory of MNF evanescent field was carried out.The physical and mathematical model was established.The principle of evanescent field sensing to detect the moisture and dissolved acetylene in transformer oil were analyzed.The four factors that affect the distribution of evanescent field were studied,including transmitted light wavelength,environmental refractive index,MNF refractive index and MNF diameter.The effect of MNF diameter on the evanescent field was explored,which laid a theoretical foundation for the selection of MNF parameters in subsequent experiments.Then,the research on the preparation of MNF was carried out.MNF melting platform was set up,and the MNF was manufactured.By changing parameters such as heater type,flame size,flame moving speed,taper speed and taper length,MNF with different diameters were prepared.The MNF had good surface smoothness and optical power transmission loss was relatively low.After that,the experiment of moisture detection in transformer oil was carried out.Oil samples of different moisture and oil chamber were prepared.MNF with different diameters were tested.When diameter was less than 40 μm,moisture detection could be achieved.As diameter decreased,detection sensitivity increased rapidly.When diameter was 800 nm,the detection sensitivity reached 1.8 μL / L.Finally,the experiment of acetylene dissolved in transformer oil detection was conducted.Oil samples with different degrees of degradation were prepared,and LIA systems were added to improve detection sensitivity.Spectral scanning parameters and data acquisition programs were designed.MNF with the diameter of 800 nm were tested.Wavelet was used to filter and denoise spectral data.The lower limit of detection reached 17.9 μL / L.