ABSTRACT
The method using ethanol to evaluate the cellulose insulation aging condition of oil-immersed transformers has been proposed. At present, the dominating method for detecting ethanol in insulating oil is to use headspace-gas-chromatography-mass-spectrometry (HS-GC-MS). However, the problem of quantitative inaccuracy will be sometimes encountered in the actual detection process due to improper instrument parameter setting and improper manual operation. In this study, as an aging marker, ethanol in transformer insulating oil was separated by using VF-624 ms capillary column. The effects of gas-chromatography-mass-spectrometry (GC-MS) optimization conditions, headspace equilibrium temperature, headspace equilibrium time and standard solution preparation method on the determination of ethanol content in oil were discussed, and optimized measures were proposed. The experimental results showed that the measurement can be more accurate under the headspace temperature of 80 °C and the headspace time of 40 min, and relative standard deviation percentage (RSD%) could reach to 4.62% under this condition. It was also pointed out that, for the preparation of standard solution, the method which controlled the sampling volume of anhydrous ethanol by microliter syringe could make the peak area of ethanol chromatogram have a better linear relationship with the standard curve. Under the similar linear range, the goodness of fitting curve without diluting process could be as high as 0.9993, while the method of preparing the stock solution and diluting stepwise to obtain the fitting curve only had a goodness of 0.9910. The method was validated by standard addition recovery test, and the recovery values obtained were between 90.3% and 95.8%. The optimized method is of great significance for the measurement of ethanol dissolved in insulating oil.
ABSTRACT
A new fingerprint compression algorithm based on sparse representation is introduced. Obtaining an overcomplete dictionary from a set of fingerprint patches allows us to represent them as a sparse linear combination of dictionary atoms. In the algorithm, we first construct a dictionary for predefined fingerprint image patches. For a new given fingerprint images, represent its patches according to the dictionary by computing l(0)-minimization and then quantize and encode the representation. In this paper, we consider the effect of various factors on compression results. Three groups of fingerprint images are tested. The experiments demonstrate that our algorithm is efficient compared with several competing compression techniques (JPEG, JPEG 2000, and WSQ), especially at high compression ratios. The experiments also illustrate that the proposed algorithm is robust to extract minutiae.
