Advanced Search

ISSN1001-3806 CN51-1125/TN Map

Volume 42 Issue 4
May  2018
Article Contents
Turn off MathJax

Citation:

Research of methane volume fraction field reconstruction based on tunable diode laser absorption spectroscopy detection technology

  • Corresponding author: CAO Shuya, 814576180@qq.com
  • Received Date: 2017-12-04
    Accepted Date: 2018-02-08
  • In order to realize 2-D distribution reconstruction of methane volume fraction field, based on tunable diode laser absorption spectroscopy (TDLAS) detection technology, methane was used as the target gas and the absorption spectrum signal of methane and nitrogen mixture was detected by direct absorption method. Volume fraction of methane was simulated and reconstructed by algebraic reconstruction algorithm. During simulation reconstruction, square reconstruction area with total 36 squares of 6×6 was used. By assuming volume fraction of a hole in a square area and simulating 24 beams through the reconstruction region from 4 directions, the projection values under the simulated ray were obtained. The results show that root mean square error of the repeated experiment is 2.58%. After adding Gaussian white noise with different signal-to-noise ratio (5%, 10%, 20%) to the simulated projection signal, root mean square error is between 4.17%~9.30%. Surface source leakage diffusion was adopted in the experiments. Artificial volume fraction hole was created and inhomogeneous volume fraction field was formed after quartz a column was placed near the center. By comparing the reconstruction results before and after the placement of quartz column, it can be seen that there is significant decrease of volume fraction in hole position. It is feasible to detect local distribution of gas volume fraction by TDLAS technology and computer tomography, and has the potential to detect volume fraction of toxic and harmful gas clouds.
  • 加载中
  • [1]

    LAI S L, WANG H, GONG P, et al. Research of propane detecting based on near-infrared laser spectral absorption[J]. Laser Technology, 2017, 41(2):284-288(in Chinese).
    [2]

    KRISHNA Y, O'BYRNE S. Tunable diode laser absorption spectroscopy as a flow diagnostic tool:A reviewreview[J]. Journal of the Indian Institute of Science, 2016, 96(1):17-28.
    [3]

    NADIR Z, BROWN M S, COMER M L, et al. Gaussian mixture prior models for imaging of flow cross sections from sparse hyperspectral measurements[C/OL].(2015-12-16)[2018-02-01].https://www.researchgate.net/publication/300413419_Gaussian_mixture_prior_models_for_imaging_of_flow_cross_sections_from_sparse_hyperspectral_measurements.
    [4]

    LINDSTROM C, TAM C J, DAVIS D, et al. Diode laser absorption tomography of two dimensional supersonic flow[C/OL].(2007-07-11)[2018-02-01].https://www.researchgate.net/publication/268482107_Diode_Laser_Absorption_Tomography_of_Two_Dimensional_Supersonic_Flow.
    [5]

    MASTERSON B, SAPPEY A, OWENBY D, et al. Flight-ready TDLAS combustion sensor for hypersonics[C/OL].(2007-07-11)[2018-02-01].https://www.researchgate.net/publication/271366737_Flight-Ready_TDLAS_Combustion_Sensor_for_Hypersonics.
    [6]

    BRYNER E, BUSA K, MCDANIEL J, et al. Spatially resolved temperature and water vapor concentration distributions in a flat flame burner by tunable diode laser absorption tomography[C/OL].(2011-01-07)[2018-02-01]. http://www.researchgate.net/publication/268566152_Spatially_Resolved_Temperature_and_Water_Vapor_Concentration_Distributions_in_a_Flat_Flame_Burner_by_Tunable_Diode_Laser_Absorption_Tomography.
    [7]

    DONG L, TITTEL F K, LI C, et al. Compact TDLAS based sensor design using interband cascade lasers for mid-IR trace gas sensing[J]. Optics Express, 2016, 24(6):A528-A535. doi: 10.1364/OE.24.00A528
    [8]

    DING N, ZHANG S B. Water vapor tomography algebraic reconstruction algorithm based on the prime number decomposition access order[J]. Geography and Geo-Information Science, 2017, 112(3):42-47.
    [9]

    JIANG Zh Sh, WANG F, XU T, et al. Study of methane telemetry based on tunable semiconductor laser absorption spectroscopy[J]. Energy Engineering, 2012(3):5-9(in Chinese).
    [10]

    MA L, LI X, CAI W, et al. Selection of multiple optimal absorption transitions for nonuniform temperature sensing[J]. Applied Spectroscopy, 2010, 64(11):1274-1282. doi: 10.1366/000370210793335052
    [11]

    BUSA K M, RICE B, MCDANIEL J C, et al. Direct measurement of combustion efficiency of a dual-mode scramjet via TDLAT and SPIV (Invited)[C/OL].(2015-01-09)[2018-02-01].http://www.researchgate.net/publication/306033342_Direct_Measurement_of_Combustion_Efficiency_of_a_Dual-Mode_Scramjet_via_TDLAT_and_SPIV_Invited.
    [12]

    LIU C, XU L, CHEN J, et al. Development of a fan-beam TDLAS-based tomographic sensor for rapid imaging of temperature and gas concentration[J]. Optics Express, 2015, 23(17):22494. doi: 10.1364/OE.23.022494
    [13]

    LI J Y. Study on temperature measurement of combustion field based on tunable laser absorption spectroscopy[D]. Tianjin: Tianjin University, 2013: 1-135(in Chinese).
    [14]

    SONG J L, HONG Y J, WANG G Y, et al. Study on two-dimensional distribution reconstruction of gas temperature and concentration in combustion field based on laser absorption spectroscopy[J]. Acta Physica Sinica, 2012, 61(24):240702(in Chinese).
    [15]

    CHENG L H. Study of iterative tomography algorithm for gas diffusion and distribution reconstruction based on incomplete projection data[D]. Hefei: Hefei Polytechnic University, 2015: 1-78(in Chin-ese).
    [16]

    XIA H H, KAN R F, LIU J G, et al. Analysis of algebraic reconstruction technique for accurate imaging of gas temperature and concentration based on tunable diode laser absorption spectroscopy[J]. Chinese Physics, 2016, B25(6):246-253.
    [17]

    BUSA K M, BROWN M S, GRUBER M, et al. Common-path measurement of H2O, CO, and CO2 via TDLAS for combustion progress in a hydrocarbon-fueled scramjet[C/OL].(2016-01-08)[2018-02-01].http://www.researchgate.net/publication/309339754_Common-Path_Measurement_of_H2O_CO_and_CO2_via_TDLAS_for_Combustion_Progress_in_a_Hydrocarbon-Fueled_Scramjet.
    [18]

    XIA H, XU Z, KAN R, et al. Numerical study of two-dimensional water vapor concentration and temperature distribution of combustion zones using tunable diode laser absorption tomography[J]. Infrared Physics & Technology, 2015, 72:170-178.
    [19]

    ZHANG G, LIU J, XU Z, et al. Characterization of temperature non-uniformity over a premixed CH4-air flame based on line-of-sight TDLAS[J]. Applied Physics, 2016, B122(1):3-11.
    [20]

    JACKSON K, GRUBER M, BARHORST T. The HIFiRE flight 2 experiment: an overview and status update[C/OL].(2009-08-05)[2018-02-01]. http://www.researchgate.net/publication/268482795_The_HIFiRE_Flight_2_Experiment_An_Overview_and_Status_Update.
  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Figures(7) / Tables(1)

Article views(4558) PDF downloads(157) Cited by()

Proportional views

Research of methane volume fraction field reconstruction based on tunable diode laser absorption spectroscopy detection technology

    Corresponding author: CAO Shuya, 814576180@qq.com
  • State Key Laboratory of Nuclear Biological Chemical Protection National, Research Institute of Chemical Defense, Academy of Military Science, Beijing 102205, China

Abstract: In order to realize 2-D distribution reconstruction of methane volume fraction field, based on tunable diode laser absorption spectroscopy (TDLAS) detection technology, methane was used as the target gas and the absorption spectrum signal of methane and nitrogen mixture was detected by direct absorption method. Volume fraction of methane was simulated and reconstructed by algebraic reconstruction algorithm. During simulation reconstruction, square reconstruction area with total 36 squares of 6×6 was used. By assuming volume fraction of a hole in a square area and simulating 24 beams through the reconstruction region from 4 directions, the projection values under the simulated ray were obtained. The results show that root mean square error of the repeated experiment is 2.58%. After adding Gaussian white noise with different signal-to-noise ratio (5%, 10%, 20%) to the simulated projection signal, root mean square error is between 4.17%~9.30%. Surface source leakage diffusion was adopted in the experiments. Artificial volume fraction hole was created and inhomogeneous volume fraction field was formed after quartz a column was placed near the center. By comparing the reconstruction results before and after the placement of quartz column, it can be seen that there is significant decrease of volume fraction in hole position. It is feasible to detect local distribution of gas volume fraction by TDLAS technology and computer tomography, and has the potential to detect volume fraction of toxic and harmful gas clouds.

引言
  • 采用光学方法进行气体体积分数检测具有速度快、精度高、非接触等优点[1-2]。基于这些优点,目前可调谐半导体激光吸收光谱(tunable diode laser absorption spectroscopy,TDLAS)检测技术在发动机尾焰燃烧诊断[3-6]、痕量气体体积分数检测[7]、有毒有害气体遥测[8]等方面已经开展了大量研究和应用。有毒有害气体的来源通常是化工厂、化工仓库事故过程中形成的有毒有害气体云团。云团内部的有毒有害气体体积分数分布是确定救援方案的重要参考,TDLAS检测技术和计算机断层重建(computed tomography,CT)技术结合为气体体积分数分布远程确定提供了一个研究方向。

    目前在气体检测领域,TDLAS检测一般采用波长调制或二次谐波进行检测。利用可调谐二极管输出波长在一定范围内可调的特点,能够同时分析多种污染物质,包括甲烷、一氧化碳、二氧化碳、一氧化氮、氨气等有毒有害气体。浙江大学JIANG等人[9]对基于TDLAS技术的望远镜收发系统的设计原理进行了研究,通过测量吸收池内甲烷标准气进行标定,并在单光程40m的遥测距离下验证了该系统的良好时间响应性。美国克莱蒙森大学MA等人在2008年提出了高光谱断层重建(hyper-spectral tomography, HT)技术[10],采用法布里-珀罗型光纤滤波激光器(fiber Fabry-Perot tunable filter laser, FFP-TFL),布置6条瞄准线(line of sight, LOS)测量路径并选择合适的吸收峰位测量了平面火焰炉的温度分布和H2O体积分数场,测量结果与相干反斯托克斯-喇曼光谱(coherent anti-Stokes Raman spectrum, CARS)的测量结果吻合良好。2011年, 该小组对HT方法进行了更深入的研究,采用15×15的分布式测量,将TDLAS系统安装在燃烧室出口处,获得温度、压力和水蒸气体积分数及这些变量的空间2维分布信息[11]。LIU等人[12]在对燃烧过程的温度进行检测的同时,也对产生的气体体积分数进行了实时测量。采用基于调制光谱的分频多路技术,对多台激光器采用不同频率的正弦电流信号进行调制,最终获得多组分的体积分数信息。天津大学的LI[13]利用TDLAS非接触测量的优点,将TDLAS双线测温技术用于高速燃烧流场的测量,选取1391.67cm-1和1468.89cm-1两条水蒸气的谱线,基于两台激光器建立了时分复用的测量系统,发展了一种基于TDLAS正交路径的2维燃烧场温度重建算法,研究了不同LOS测量路径数对单峰温度场和非对称双峰温度场的2维重建结果的影响,最后建立并验证了快速温度宽谱调谐多谱线测温体系。装备学院的SONG等人[14]采用非规则光线布局,提出了一种非规则光线分布优化原则,基于代数重建算法(algebraic reconstruction algorithm, ART)进行温度场2维分布重建。研究了发射端数目对重建结果的影响,通过实验验证得出了与仿真结果相吻合的结论。合肥工业大学的CHENG等人[15]将迭代重建算法用于不完全投影数据的重建工作中,针对光学遥感技术获取数据通常不完整的特点,对代数重建算法、乘型代数重建算法、联合代数重建算法、改进的联合代数重建算法等算法进行了改进,实验结果表明,改进算法在抗噪性方面有极大改善。中国科学院安徽光学精密机械研究所的XIA[16]等人使用改进的ART代数重建算法,对TDLAS检测结果进行了断层重建,获取了平面火焰炉燃烧产生水汽的2维分布重建结果。

    痕量气体检测和燃烧诊断领域国内外已有大量研究,但对于有害气体的体积分数场分布测量研究较少,TDLAS技术本身的高分辨率特点决定了通常只需更换激光器即可完成不同气体的检测任务,因此将TDLAS检测技术用于有毒有害气体检测领域是完全可行的。本文中以常见的有害气体甲烷为研究目标,采用TDLAS技术进行非接触式体积分数测量,结合CT技术重建气体的体积分数场分布。实验中采用旋转台来实现对气体喷发区域的多角度扫描,获得24条光线数据,应用ART算法,来实现对6×6网格范围内甲烷气体的2维体积分数分布重建。

1.   测量原理
  • 本文中主要采用TDLAS气体测量原理和代数重建算法来实现对甲烷扩散区域的2维体积分数分布重建。

  • TDLAS气体测量应用朗伯比尔定律[16]:

    在单一跃迁前提下可以写作:

    式中, It为透射光强(mW),I0为入射光强(mW);kν为光谱吸收系数(cm-1),由气体的静态压强p(Pa)、线强S(T)(cm-2·Pa-1)、线型函数φν和体积分数x确定;ν为入射光频率(cm-1),L为有效吸收光程(cm),φν为归一化的线型函数,其积分值为1。

    对(2)式进行积分,进而忽略线型函数的影响,可以得到积分吸光度A和体积分数x之间的关系,即:

    得到一个包含积分吸光度A、体积分数x、光程L的方程组。从而可以利用ART方法进行迭代计算。

  • ART代数重建算法是迭代重建法的一种形式,它的特点是:预设一初始图像x0,以x0为基础计算x1,进而根据x1计算x2,重复迭代至满足预设收敛条件,而后终止。每一次通过xk计算xk+1时,只需考虑一条光线产生的校正值Δxk,其所修正的数据网格也只包含这条光线通过的网格[17-20]。其本质上是通过一条条光线对所选初值的迭代修正,逐步令重建结果不断逼近实际值,在迭代修正过程中每一条光线并不总能起到正面的作用,在实验数据不够稳定的情况下,加入一些误差较大的光线将会放大误差,降低重建效果,对于重建算法中光线的选取是一个比较重要的环节,并且与实际测量的体积分数场分布及光线分布有着紧密的联系。

    将(2)式改写为:

    式中,A为实验数据处理后获得的积分吸光度向量,x为所求体积分数向量,而矩阵R则为投影系数矩阵。

    采用ARTⅡ算法,则有:

    式中,ik=k(modI)+1,k代表迭代次数,IR的维数,λ为松弛因子,且取值在(0,2]之间,aik为积分吸光度A的分量,rik为投影系数矩阵R的分量,经过k次迭代满足截止条件时的xk即为重建结果。算法流程图如图 1所示。

    Figure 1.  ART flow chart

2.   仿真实验
  • 从方便计算投影系数矩阵的需求出发,加快运算速度,仿真实验采用方形区域进行重建,区域分划为6×6,选取水平垂直方向各6道光线射入,45°及135°方向斜向选取各6条光线,总共有24条光线穿过投影区域。如图 2所示。

    Figure 2.  Sketch of beam distribution

    Figure 3.  Simulate distribution of preset volume fraction

    预设体积分数分布如图 3所示。边缘略低于中心区域,中心点的右上部分为一空穴,重建结果如图 4所示,迭代次数上限设置为100000次,并以均方根误差变化差值不超过10-14为迭代终止条件。经过重复实验统计均方根误差在2.58%,对模拟投影信号加入不同占空比的高斯白噪声之后再进行重建,均方根误差分别在4.17%~9.30%之间。

    Figure 4.  Results of simulation reconstruction

3.   实验结果
  • 本实验中采用如图 5所示装置。将用于模拟甲烷泄漏的平面炉放置在旋转台的中心,使用电磁流量计控制气流速度,在旋转台上正交方向布置两组光线准直器及探测器,每组6对,通过旋转至0°, 45°, 90°, 135°分别采样,共获取24路信号,激光器发出的光经由分束器连接到各个准直头,其中一路用于同步定标。实验设计在平面炉的中心点右上放置透光的石英柱,对比是否放置石英柱的重建结果来与仿真实验相互验证。

    Figure 5.  Experimental installation sketch

    实验中使用Nanoplus公司的分布式反馈激光器(distributed feedback laser,DFB),其中心波长覆盖到甲烷吸收峰位1653.72nm,在常温下吸收强度良好。对采集到的信号进行处理,计算出,经过查表获取常温下的线强值为1.097×10-8cm-2·Pa-1,压强p=1.01×105Pa,可以计算出的值,然后将其作为迭代信息进行迭代重建。

    TDLAS信号检测采用直接测量法,获取信号如图 6所示。选取信号差异区域(长约500个点)前后各1000个点,利用干涉定标信号确定横坐标,然后计算积分吸光度A

    Figure 6.  TDLAS signal detected by direct measurement method

    重建结果如表 1图 7所示。表 1中上下两部分的6×6数据表格分别对应图 7a图 7b各网格体积分数值。

    volume fraction of methane
    0 0.0002613 0.002157 0 0.003408 0
    volume fraction distribution of Fig. 7a 1.199×10-6 0.01037 0 0.003102 1.638×10-6 0
    0.005740 0 0.01137 0.0006311 0 0
    1.382×10-6 0.0009636 0.006911 0 0.0006864 0.002332
    3.653×10-6 0.004569 0 0.002429 0.0001081 0.001578
    0.005721 3.653×10-6 1.382×10-6 0 0.001555 0.0004072
    volume fraction of methane
    0.0002642 0.0002642 0.0002642 0.007122 0.01247 0.007085
    volume fraction distribution of Fig. 7b 0.01339 0 0 0.006233 0.006762 0
    0.003168 0.01182 0.01704 0.01841 0 0.004573
    0.01162 0.01373 0.004707 0 0 0.005895
    0.005415 3.391×10-5 0 0.0003252 0.008037 0
    0 3.610×10-5 0.01378 9.827×10-5 0 0

    Table 1.  Volume fraction distribution of Fig. 7a and Fig. 7b

    Figure 7.  Gradient graph of reconstruction results

    图 7中的重建结果可以看出,平面炉出气并不均匀。图 7a图 7b中已用方框标注出石英柱摆放区域。从图 7b图 7d可以看出,在放置石英柱之前,甲烷主要集中在中心区域附近。从图 7a图 7c可以看出,放置石英柱挡住了右侧之后,重建结果显示出右侧出现空穴,并对周边甲烷体积分数产生了影响,降低了右侧区域整体体积分数,左侧区域体积分数变化不大,与实验前预测相吻合。实验中遇到的主要问题在于气流的不稳定性,通过大量的重复试验来获取平均数据并不能完全克服这一问题,需采取增加单次扫描光线数量和提高气体体积分数缓和误差的方法改进实验,在考虑成本的前提下提高测量的准确度和重复性。

4.   结论
  • 基于可调谐半导体激光吸收光谱技术,通过快速旋转台获取不同角度下的光线信息,使用ART代数重建算法对存在空穴的体积分数场进行重建,使用单激光器达成了这一目标,通过仿真实验和实际测量相互验证,为之后对气体体积分数场的快速3维重建打下基础。单激光器分束配合旋转台的实验方式硬件成本较低,与之相对的实验误差及时间成本也较多激光器分束、固定位多角度同时测量方法更大,对于变化较快的气体体积分数场而言,应用固定位多角度测量方式改进实验装置,提高数据时效性。

Reference (20)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return