为了实现烟羽二维空间分布的精确重建,建立了被动多轴差分吸收光谱断层扫描系统,实现了多台被动多轴差分吸收光谱系统对废气烟羽空间分布的测量.首先,介绍了多轴差分吸收光谱系统及其反演气体浓度的机理.接着,使用代数迭代算法对不同的重建模型,采用不同的扫描光路进行了重建模拟,并设计了重建程序.然后,对数值模拟仿真结果进行了分析比较.最后,搭建了多轴差分吸收光扫描系统平台,进行了外场试验.数值模拟的结果显示:MAX-DOAS层析技术能精确的重建出烟羽模型的二维空间分布,四光源光路的重建误差约是双光源光路重建误差的三分之一,四光源重建时间约是双光源重建时间的四分之一,且双峰模型的重建误差大于单峰模型的重建误差.外场试验的结果显示:重建图像的积分数据与多轴差分吸收光谱测得的投影数据是一致的,说明重建出的烟羽空间分布符合实际情况.研究表明,数值模拟的结果与外场试验的结果具有一致性.

In order to achieve precisely two-dimensional spatial distribution reconstruction of smoke plume,passive MAX-DOAS tomography is established,the measurement of the spatial distribution of the exhaust plume is implemented by more passive multi-axis differential absorption spectrum system.First,the multi-axis differential absorption spectrum system and its mechanism of inverse gas concentration are introduced in the paper.Then,algebra iterative algorithm is adopted to extract the information of the trace gas concentration in reconstruction simulation with different models and different scanning optical path,and the reconstruction program is designed.Then,the numerical simulation results are compared.Finally,a platform of multi-axis differential absorption optical tomography system is set up,a field campaign was carried out.The numerical simulation results show that the MAX-DOAS tomography can accurately reconstruct two-dimensional spatial distribution of plume model,the reconstruction error of MAX-DOAS tomography with four light sources is about a third of the reconstruction error with double light sources,moreover,the reconstruction time is about a quarter of the reconstruction time of double light sources,and the reconstruction error of the twin peaks model is greater than that of the one peak model.Field test results show that the integral data of reconstruction image is consistent with the measured projection data of multi-axis differential absorption spectrum,the spatial distribution reconstruction of plume is in line with the actual situation.Studies have shown that the result of numerical simulation and field test results have consistency.