能源高效清洁利用和尖端装备快速发展对热工过程气体参数的精确测量提出了更高的要求。为解决可调谐二极管激光吸收光谱技术（TDLAS）中直接吸收法抗干扰能力差、高灵敏度波长调制法无法精确测量分子吸收截面的问题，本项研究拟基于FFT频谱分析理论，在固定点波长调制策略下，推导出蕴含分子吸收截面信息、具有较强抗干扰能力的透射光强-频率模型；在该模型基础上，充分考虑激光光强和频率非线性效应，推导出无吸收时基线-频率高精度表达式，结合同步拟合思想，建立一种高精度、免标定的分子吸收截面在线测量方法；最后，以CO和CO2气体为例，利用高温管式炉模拟多种温度、压力实验工况，从吸收截面的拟合残差、气体浓度和谱线参数的测量精度以及艾伦方差等方面对该方法的抗干扰能力、精度和测量下限进行评价。该方法融合直接吸收法和波长调制法的优点，预期可为热力系统优化控制、燃烧诊断和复杂工业现场痕量气体浓度监测提供新的测量方法。

With the requirement of efficient and clean utilization of energy and the development of cutting-edge equipment, more demands are put forward to the high-precision measurement of gas parameters. The tunable diode laser absorption spectroscopy (TDLAS), as the most potential candidate, encounters the following problems: the direct absorption spectroscopy (DAS) is susceptible to noise and the wavelength modulation spectroscopy (WMS) is unable to measure the important absorbance line profile. To solve these problems, in this project, a transmitted light intensity model with high noise resistance will be first built based on the FFT spectrum analysis in the fixed wavelength modulation scheme. Based on this model, a high-precision baseline expression will be derived in full consideration of the non-linearity of laser intensity and wavelength. Then, combining the simultaneous fitting idea, an in-situ, calibration-free method to measure the absorption line profile will be established. Finally, the proposed method will be experimentally validated in a high-temperature tube furnace taking CO and CO2 as examples. Multiple indicators, including the fitting residual of absorbance line profile, the measurement accuracy of gas concentration and spectral line parameters will be used to verify the performance and accuracy of the proposed method. The detection limits will also be further checked with the Allen variance. The proposed method, which combines the advantages of DAS and WMS, is expected to be a preferable measurement approach for the optimal control of thermal system, combustion diagnostics, and trace gas concentration monitoring in complex industrial scenarios.