The Theoretical Analysis And Experimental Researches Of The Mineral Oils Based On Fluorescence Detection

With the rapid development of China’s economy, the demands for energy areconstantly increasing, this promoting the ocean petroleum industry and transportationindustry’s booming. This situation leads the situation of our country ’s offshore oilpollution directly to be getting more and more serious. Over the years, more and moreleaked incidents of oil pollution that are caused by unknown source have occurredfrequently in the China Sea. Especially in the last two years, pollution accidents aregreatly increased.Thus, marine oil pollution in our case is needed to be solved urgently. In the processof marine pollution detection, the rapid and accurate determination of oil can quicklyindicate the direction of the investigation for the accident investigators, as much aspossible to reduce the scope and period of the investigation and make the investigationefficient, which is important for the successful completion of the maritime oil spill’sinvestigation which is caused by unknown source.The paper, using the fluorescence spectroscopy as the main method and based on thedeep analysis of the characteristics of oil pollutant composition, proposes the newmeasurement methods of the petroleum contaminants’ concentration detected and speciesidentification which is based on the combination of fluorescence spectroscopy,multivariate statistical analysis and the dimension reduction of three-dimensional images,and make the qualitative and quantitative analysis of specific fluorescent mineral oil tocomplete the experimental demonstration of this new method.The methods of dimensionality reduction and envelop processing are firstlyintroduced to the detection of mineral oil, which greatly simplifies the quantity ofexperimental fluorescence data; through the study of the apparent features of two andthree dimensional spectrum, the paper chooses the best features extracted, which arestandard error of the mean, kurtosis coefficient and skewness coefficient; and through avariety of study of the contrast experiments using multivariate statistical analysis method,it proves that the system clustering method can realize the distinction of oil well, whose accuracy is above98%, while comparing with parallel factor analysis method, theresolution efficiency of time is improved by more than30%.As for detection for mineral oil content in water, the paper can realize the content’saccurate measurements in the optimal excitation and emission wavelength selectionproved by experiment, and both the linear relationship and the recovery rate are more than96%.