非常规油浓度分辨荧光光谱多环芳烃的风化特性研究  

Study on weathering characteristic of aromatic hydrocarbons specific to unconditional oil based on concentration-resolved-fluorescence spectroscopy

作　　者：王春艳[1,2] 黄小东 樊心民[1] 杨春[2] 王镇棣 WANG ChunYan;HUANG XiaoDong;FAN XinMinl;YANG Chun;WANG ZhenDi(Department of Physics and Optoelectronic Engineering, Weifang University, Weifang 261061, China;Emergencies Science and Technology Section （ESTS）, Science and Technology Branch, Environment Canada, Ottawa K1A OH3, Canada)

机构地区：[1]潍坊学院物理与光电工程学院,潍坊261061 [2]Emergencies Science and Technology Section (ESTS), Science and Technology Branch, Environment Canada

出　　处：《中国科学：技术科学》2018年第5期548-557,共10页Scientia Sinica(Technologica)

基　　金：山东省自然科学基金英才计划(编号:ZR2015M007);山东省自然科学基金培养基金(编号:ZR2018PF016);国家自然科学基金(批准号:40706037;41476081);山东省博士基金(编号:BS2011HZ015);山东省重点研发计划(编号:2016GSF115020)资助

摘　　要：随着常规原油的开采量和运输量的迅速减少,重质原油、稠油、油砂、油页岩等非常规油逐渐成为海洋溢油的主体,以往主要针对常规原油检测的技术方法遇到了新的挑战.依据时间分辨荧光光谱的思路,将浓度作为新的一维坐标引入荧光光谱,发展了浓度分辨荧光光谱(concentration-resolved-fluorescence spectroscopy,CRFS),对不同原油样品模拟风化实验、高温馏分实验,结果表明该方法具有很强的抗风化抗干扰能力.并对相似稠油原油样品进行了120 d的光氧化及海水风化实验研究,利用二维Gabor小波分析对视为图像信息的荧光光谱进行特征提取,结合支持向量机(support vector machine,SVM)对风化样本进行分类识别,准确率达96%,该方法快速、准确、经济,特别适合稠油等非常规溢油的检测,有望成为溢油排查的重要技术,也可成为油气地球化学研究的有效工具之一.With the rapid decline of conventional crude oil＇s production and transportation, heavy crude oil, heavy oil, oil sands, oil shale and other unconventional oil have gradually become the main body of the marine oil spills. The used oil spill fingerprinting methods mainly for conventional oil are facing new challenges. According to the time-resolved fluorescence spectroscopy, concentration value is introduced as a new data dimension to the fluorescence spectra and the concentration-resolved-fluorescence spectroscopy（CRFS）is developed to express more PAHs information. Laboratory simulation experiments of weathering and distillation were carried out and the results demonstrate this method has strong anti-weathering and anti-interference ability. Oil source classification of closerelated heavy crude oil samples under light weathering and seawater weathering lasting 120 d are studied. 96% of the correct rate for the closely related oil source samples is achieved by combining Gabor wavelet with supported vector machine（SVM）. The obtained results suggest that the newly-developed method may become specifically applicable in spilled oils identification and also can be applied to other multi-fluorophoric systems such as in petroleum exploration, dissolved organic matter in environment monitoring and petroleum exploration.

关 键 词：溢油鉴别 荧光光谱 风化 模拟实验 多环芳烃 

分 类 号：O657.3[理学—分析化学] X834[理学—化学]