Photoacoustic spectroscopy detection and extraction of discharge feature gases in transformer oil based on 1.5 μ tunable fiber laser

Information from the analysis of gasses dissolved in insulating oils is valuable for early a transformer maintenance. By means of dissolved gas analysis (DGA), it is possible to distinguish faults such as partial discharge (corona), overheating (pyrolysis) and arcing in a great variety of oil-filled equipment. Tunable fiber laser-based second harmonic photoacoustic spectroscopy offers a fast and good-noise-immunity technique for the quantitative analysis of trace gases in transformer oil. In this work, the discharge feature gases, such as C2H2, CH4, CO2 and H2O, were measured with a tunable laser photoacoustic spectrometer at the 1530.3709 nm transition line, as a typical application of precise measurement of multi-gas, a proposed BSS model based on overcomplete ICA basis and five-point-sampling method based on a created weight-truncation-constraint equation was used to remove noise so that several fault gases can be extracted with a higher detection accuracy and a method detection limit. Experiment shows that within 0.15 nm band near 1530.3709 nm, the four-gas contents have been detected and extracted and the detection accuracy has been improved from available scanning spacing of 0.03 nm to available extracting spacing of 0.0011 nm. At room temperature and atmospheric pressure, this can achieve simultaneous detection for multiple feature gases in discharged transformer oil using laser source with a limited waveband.