To realize the on-line fluorescence monitoring of mineral oil pollution in water, three-dimensional spectral characteristic of oil-water intermixtures must be studied and the characteristic must be extracted. Using excitation wavelength, fluorescence wavelength and fluorescence intensity as three-dimensional system of coordinate, through sampling and surface fitting, three-dimensional fluorogram is gotten, which can provide gist for oil discrimination when presented in contour chart (finger-print map of oils). But there is little difference between characteristics of three-dimensional fluorogram because of the similarity of constituent and structure of similar oils. Therefore this paper introduces quantitative analysis method-characteristic parameter method which starts with analyzing statistical characteristic of three-dimensional fluorogram. Using RFPC fluorescence spectrometer (Shimadzu, Japan), three-dimensional fluorescence spectra of diesel oil, machine oil, gasoline oil, crude oil are measured and parameterized. The result shows that as a quantitative classified discrimination method of three-dimensional fluorescence spectra, the parameter of characteristic parameter method possesses definiteness for three-dimensional fluorescence spectra, and it is applicable, available when used in oil discrimination.
Recently, chromatography column and gas sensor have been used in online monitoring device of dissolved gases in transformer oil. But some disadvantages still exist in these devices: consumption of carrier gas, requirement of calibration, etc. Since FTIR has high accuracy, consume no carrier gas and require no calibration, the researcher studied the application of FTIR in such monitoring device. Experiments of “Flow gas method” were designed, and spectrum of mixture composed of different gases was collected with A BOMEM MB104 FTIR Spectrometer. A key question in the application of FTIR is that: the absorbance spectrum of 3 fault key gases, including C2H4, CH4 and C2H6, are overlapped seriously at 2700~3400cm-1. Because Absorbance Law is no longer appropriate, a nonlinear calibration model based on BP ANN was setup to in the quantitative analysis. The height absorbance of C2H4, CH4 and C2H6 were adopted as quantitative feature, and all the data were normalized before training the ANN. Computing results show that the calibration model can effectively eliminate the cross disturbance to measurement.
Fourier Transform Infrared gas analyzers have been widely used for speedy quantitative analyses of gases, and it is found that in many cases field maintainability determines the instruments’ online applicability instead of the instrument’s accuracy as is desired. To be maintenance-free is both the target of online instruments and the key to their field applications. Analyses show that if a background can be collected simultaneously with the sample spectrum, the transmittance will be only a function of concentration. Gas spectra collected on Nicolet670 of attenuated inputs and adjusted gains are examined via OMNIC software. Collected data exhibit that the absorbance spectra keep constant when input energy increases 8 times and the instrument gains becomes to 2.0 and 4.0 times. On viewing the absorption peaks vary with wavenumber in high “frequency” and that of the spectrometer itself in low frequency, by subtracting its instrumental response function of a transmittance spectrum extracted by wavelet transform, an absorbance spectrum of air is obtained and comparisons demonstrate that it is a summation of absorbance of the sample and that of gases in the optical path of the spectrometer. Calibration-free and background-free principles are thus exhibited and they construct maintenance-free principles of FTIR gas analyzers.
FTIR spectra have been widely used in quantitative analyses of mixed gases. Although Beer's Law regulates the relationships between absorbency and the product of concentration and path length, its deviations have been found rather complicated. Here we present the complexity of quantitative relationships between methane's infrared spectra and concentrations and resolutions. Measurements of the same methane sample's spectra under different resolutions demonstrate that both area absorbency and height absorbency vary with resolutions; spectra at lower resolution have bigger area absorbency for most of the peaks and are more likely to saturate for peaks of strong absorption. Standard methane sample of certain concentration is diluted with super pure nitrogen via mass control flowmeters and continuously passes through a 2-meter gas cell, such that the spectra of methane of different concentrations are collected. The area absorbencies of different peaks are carefully calculated via OMNIC software and results show that peaks with lower absorption are more likely to fit to linearity but more reluctant to changing concentrations. Area absorbencies are integrated through characteristic absorption regions, height absorbencies and area absorbencies are calculated at the two main absorption peaks and measurements show that approximative linearity fits all the areas and the best linearity appears at 1035.6cm-1.
Dissolved Gas-in-oil Analysis is one of the most effective methods to diagnose the potential inner faults of power transformers. To overcome the disadvantages of the current sensing methods including gas chromatographic and Fourier Transform InfraRed techniques, photoacoustic spectroscopy has been introduced for both online and offline monitoring of transformer fault gases. Pulsed infrared sources are recommended and the wavelengths for each diagnostic gas are suggested. For hydrogen has no absorption in Infrared band, in no way can it be detected by infrared spectroscopy. Therefore, a novel technique is introduced to detect hydrogen via the phase shift of photoacoustic signals. The detailed sensing principle gives that the relative time shift of the photoacoustic signals by the adding of each fault gas is proportional to the added concentration of the very gas. By subtracting the effect of all the other fault gases, the added concentration of hydrogen can be calculated. Analyses show that the error of this method may be smaller than that of measuring other fault gases; the lowest sensible limits from 5ppm to 60ppm may require the time resolution of the system to be within the range of 10-6~10-9 and they are tested to be reachable by simulations on LabVIEW.
Recently, Semiconductor sensor and thermal conductivity sensor are widely used for gas detection in transformer online monitors. Since the long-time stability or precision of these sensors is not satisfactory, the present researcher studied the application of FTIR in such monitors. In the wide measuring range of online monitoring, Absorbance Law is not always applicable, thus a non-linear calibration model was necessary. Experiments were done to set up the calibration model. A gas dilution system was designed. With the system, standard samples of fault gas including CH4, C2H2, C2H4 and C2H6 were diluted to different concentration. BOMEM MB104 FTIR Spectrometer was used to collect spectra of gases. Curve fitting of the output of FTIR was done, and the effect of quantitative feature and concentration range on quantitative analysis was investigated. In addition, the lowest detection limit was tested. Experiment and calculation results show: accuracy can be improved by taking strong peak height at low concentration range, taking peak area or weak peak height at high concentration range as quantitative feature, and using third order polynomial to fit the output curve of FTIR. The lowest detecting limit of C2H2 with 2.4m gas cell is below 0.3ml/l and that of 10cm cell is below 3ml/l.
Dedicated experiments were made to clarify interference of the oil vapor, establish the quantitative analysis model and test its effectiveness in analyzing transformer fault gases with Fourier Transform Infrared. Nicolet Nexus 670 FTIR Spectrometer with a 2-meter vacuumable gas cell is used to collect and examine the spectra of the top gap gases and pyrolytic products of the transformer oil and no additional interference other than diagnostic gases were found. Standard gas samples of CH4, CO and CO2 are diluted with super pure nitrogen via mass control flowmeters to get samples with different concentrations dynamically. Spectra of the diluted samples are collected, analyzed and used as calibration standards to measure other samples. It is shown that the absorbance of the spectra can be calculated in height of the peak or in area under the peak and they vary with concentrations in approximate linearity, yet the former is more active and the latter is more passive to be saturated. Repeated measurements say the relative error of the established model is below 4% and typical measured concentrations are 537μL/L for methane and 57μL/L for monoxide.