光学被动成像干涉技术探测深海热液CH_4  

Detection of abyssal hydrothermal CH_4 based on optical passive imaging interference technology

作　　者：刘青松 胡炳樑[1] 唐远河[3] 于涛[1] 王雪霁 刘永征[1] 杨鹏 王浩轩 Liu Qingsong;Hu Bingliang;Tang Yuanhe;Yu Tao;Wang Xueji;Liu Yongzheng;Yang Peng;Wang Haoxuan(Key Laboratory of Spectral Imaging Technology,Xi′an Institute of Optics and Precision Mechanics,Chinese Academy of Sciences,Xi′an 710119,China;University of Chinese Academy of Sciences,Beijing 100049,China;School of Science,Xi′an University of Technology,Xi′an 710048,China;Joint Laboratory for Ocean Observation and Detection,Qingdao National Laboratory for Marine Science and Technology,Qingdao 266200,China)

机构地区：[1]中国科学院西安光学精密机械研究所光谱成像技术重点实验室,陕西西安710119 [2]中国科学院大学,北京100049 [3]西安理工大学理学院,陕西西安710048 [4]青岛海洋科学与技术国家实验室海洋观测与探测联合实验室,山东青岛266200

出　　处：《红外与激光工程》2018年第9期40-45,共6页Infrared and Laser Engineering

基　　金：国家重点研发计划项目(2017YFC1403700);国家自然科学基金面上项目(61675165);中国科学院光谱成像技术重点实验室开放基金(LSIT201714D)

摘　　要：为了实现对热液甲烷浓度、温度和压强信息的实时、长期探测,提出一种新颖的光学被动成像干涉系统(Optical Passive Imaging Interference System,OPIIS),并建立了该系统的正演模型和反演模型。首先利用IDL语言建立了包括深海气体辐射模型、海水传输模型和仪器响应模型的OPIIS正演模型,并模拟其正演干涉图。正演干涉图信噪比总体处于50~70,浓度探测灵敏度为0.1 mmol/L,温度灵敏度为2 K,压强灵敏度为0.1 MPa。其次采用成像干涉技术结合偏最小二乘法的方法进行OPIIS数据的精确、快速反演。利用25个建模样本建立了甲烷多因变量PLS回归模型,并利用25个预测样本对回归模型进行交叉检验。该最优回归模型的浓度预测最大误差为1.9%,温度预测最大误差为0.38%,压强预测最大误差为1.0%。An optical passive imaging interference system(OPIIS)was proposed for the real-time and long-term detection of hydrothermal methane′s concentration,temperature and pressure.Firstly,the forward model that consisted of deep ocean gas emission model,seawater transmission model and instrument responding model was built by interface description language(IDL),and its forward interference fringers were simulated.The SNRs of the forward interference fringes were in the range of(50-70)in general.And the detection sensitivity of concentration measurement is was at least 0.1 mmol/L,the temperature was at least 2 K,and the pressure was at least 0.1 MPa.Then,OPIIS′s data were processed accurately and rapidly by combining imaging interference technology and partial least squares(PLS)algorithm.The multi-dependent variable PLS regression model of methane was established by using 25 modeling samples,and this PLS regression model was cross-validated by using 25 prediction samples.And the max error for concentration prediction of this regression model was 1.9%,for temperature prediction was 0.38%,and 1.0%for pressure prediction.

关 键 词：甲烷 热液 正演 反演 

分 类 号：P716.42[天文地球—海洋科学]