基于高光谱吸收层析技术的火焰温度场测量方法研究

Temperature field is one of the most important factors that determine combustion efficiency and pollutants generation. The imaging technique based on hyperspectral absorption tomography is an effective tool that can measure the temperature field. Especially for engine measurements with limited optical access, this method can reconstruct the temperature field using just a few line-of-sight measurements. This paper aims to investigate the major topics that are relevant to this technique. The applicant quantitatively analyzed the relationship between the beam arrangement and the coefficient matrix, and proposed a beam optimization method based on the so-called relevance matrix. For applications with a limited number of line-of-sight measurements, the applicant explored a regularization method that is enlightened by compressed sensing. In addition, the applicant proposed a method for quantification of spatial resolution by measuring the edge spread function of a hyperspectral tomographic system. Based on this method, the applicant further investigated to what extent the spatial resolution can be improved by enhancing the spectral sampling. The in-depth study of these topics are beneficial for the improvement of hyperspectral absorption tomography and lays a solid ground for its potential applications to engine measurements.

燃烧过程中的温度场分布是影响燃烧效率和污染物生成最重要的因素之一。高光谱吸收层析测温技术是测量火焰温度场的有效手段，特别是对于光学窗口有限的发动机测量，该技术只需较少的测量光路即可实行温度场重建。该项目致力于研究高光谱吸收层析测温技术中的若干核心问题。针对测量光路优化布置的问题，申请人定量解析了系数矩阵和光路布置的内在联系，提出了基于相关度矩阵的测量光路优化布置方法；针对测量光路不足时的层析反演正则化问题，申请人着重探索了基于压缩感知思想的正则化方法；针对高光谱吸收层析测温系统空间分辨率的标定问题，申请人提出了基于测量边缘扩散函数的标定方法，并在此方法的基础上，进一步研究通过增加光谱维度采样的方式在多大程度上可以改善高光谱吸收层析测温系统的空间分辨率。这些问题的深入研究可为高光谱吸收层析测温技术本身的完善及其在发动机燃烧诊断中的应用奠定良好的基础。

温度场的测量对于电站锅炉、燃气轮机等动力装置的诊断和优化具有重要意义。本项目针对高光谱吸收层析技术的若干重要问题开展研究。针对高光谱吸收层析系统的光路优化问题，发展了基于正交度矩阵的优化方法，研究结果表明该方法可以有效提高温度场的重建精度；针对高光谱吸收层析反演的病态问题，重点研究了基于全变分的正则化和Tikhonov正则化，结果表明前者适用于分块平滑的温度场重建，而后者则适用于整体平滑的温度场重建；针对高光谱层析系统空间分辨率的标定问题，提出了通过测量边缘扩散函数的方式来计算点扩散函数、调制传递函数和空间分辨率的方法。项目期间共发表SCI论文24篇、国内科技核心期刊论文1篇；获得德国埃尔朗根高等光学研究院青年科学家奖；组织第330期东方科技论坛学术研讨会“计算成像及动力系统燃烧诊断”以及“第四届全国激光光谱技术学术论坛”；并受邀在美国航空航天学会2018航空论坛做邀请报告。