This is a comprehensive, critical, and pedagogical review of volumetric emission tomography for combustion processes. Many flames that are of interest to scientists and engineers are turbulent and thus inherently three-dimensional, especially in practical combustors, which often contain multiple interacting flames. Fortunately, combustion leads to the emission of light, both spontaneously and in response to laser-based stimulation. Therefore, images of a flame convey path-integrated information about the source of light, and a tomography algorithm can be used to reconstruct the spatial distribution of the light source, called emission tomography. In a carefully designed experiment, reconstructions can be post-processed using chemical kinetic, spectroscopic, and/or transport models to extract quantitative information. This information can be invaluable for benchmarking numerical solutions, and volumetric emission tomography is increasingly relied upon to paint a more complete picture of combustion than point, linear, or planar tools. Steady reductions in the cost of optical equipment and computing power, improvements in imaging technology, and advances in reconstruction algorithms have enabled a suite of three-dimensional sensors that are regularly used to characterize combustion. Four emission modalities are considered in this review: chemiluminescence, laser-induced fluorescence, passive incandescence, and laser-induced incandescence. The review covers the reconstruction algorithms, imaging models, camera calibration techniques, signal physics, instrumentation, and post-processing methods needed to conduct volumetric emission tomography and interpret the results. Limitations of each method are discussed and a survey of key applications is presented. The future of volumetric combustion diagnostics is considered, with special attention paid to the advent and promise of machine learning as well as spectrally-resolved volumetric measurement techniques.

这是对体积的全面、批判性和教学性的回顾燃烧过程的发射断层扫描。科学家和工程师感兴趣的许多火焰都是湍流的，因此本质上是三维的，特别是在实际的燃烧器中，它通常包含多个相互作用的火焰。幸运的是，燃烧会导致光的发射，无论是自发的还是响应基于激光的刺激。因此，火焰图像传达了关于光源的路径集成信息，可以使用断层扫描算法来重建光源的空间分布，称为发射断层扫描。在精心设计的实验中，可以使用化学动力学、光谱和/或传输模型对重建进行后处理，以提取定量信息。这些信息对于基准数值解决方案非常宝贵，与点、线性或平面工具相比，体积发射断层扫描越来越多地用于绘制更完整的燃烧图。光学设备和计算能力成本的稳步降低、成像技术的改进以及重建算法的进步使得一套经常用于表征燃烧的三维传感器成为可能。本综述考虑了四种发射模式：化学发光、激光诱导荧光、被动白炽灯和激光诱导白炽灯。审查涵盖了进行体积发射断层扫描和解释结果所需的重建算法、成像模型、相机校准技术、信号物理、仪器和后处理方法。讨论了每种方法的局限性，并对关键应用进行了调查。考虑了容积燃烧诊断的未来，特别关注机器学习以及光谱分辨容积测量技术的出现和前景。