可调谐二极管激光吸收光谱(TDLAS)是一种非侵入式光谱检测技术, 具有高选择性、 高响应性和高分辨率等特点。 根据分子光谱吸收原理, 被检测气体所处环境温度的改变会引起分子吸收谱线强度的变化, 进而影响气体浓度反演的准确性。 为提高气体在高温背景下浓度测量的准确性和真实性, 选取工业过程常见的一氧化碳(CO)为目标气体, 设计了基于波长调制技术多温度梯度(室温14～1 100 ℃)的气体吸收光谱检测实验, 与HITRAN数据库中光谱参数进行对比, 并对结果进行校正和分析。 同时, 以探测信号有效扫描区域的线性度、 标准差和残差平方和等参数为依据, 分析了不同材质的窗片对高温实验的影响, 通过升降温实验数据的分析, 选择了降温梯度测量作为高温实验的最佳控温顺序。 经过对标准浓度的CO进行高温实验, 发现随着温度的升高, 二次谐波(2f)幅值和吸收线强有相一致的下降趋势, 符合理论公式的变化规律。 经过分析校正后的2f幅值和温度呈现非相关性, 实现了热背景下光谱检测的校正, 验证了变温时2f幅值校正的准确性。 该研究为光谱检测技术在高温背景下实际应用提供了一定的参考, 尤其是对高精度工业炉内气体燃烧效率的动态评估具有极其重要的意义。

Tunable diode laser absorption spectroscopy (TDLAS) is a non-invasive spectral detection technology with high selectivity, high response and high resolution. According to the principle of molecular spectral absorption, the change in target gas temperature will affect the change of molecular absorption line strength and then affect the accuracy of gas concentration inversion. In order to improve the accuracy and authenticity of gas concentration measurements in high-temperature atmospheres, carbon monoxide (CO), a common gas in industrial processes, was selected as the target gas. Experiments were designed to detect the gas spectra in multiple temperature regions (from 14 to 1 100 ℃) based on wavelength modulation technique, compared with spectral parameters in the HITRAN database, and the results were calibrated and analyzed. At the same time, the influence of different materials of sapphire window pieces was analyzed in terms of parameters such as the linearity of the detection signal. A cooling gradient measurement was selected as the temperature control sequence for the high-temperature experiment by analysing the data from the temperature rise and fall experiments. Through the high-temperature experiment with a standard concentration of CO, it was found that the second harmonic (2f) amplitude and absorption line intensity had a consistent decreasing trend with increasing temperature, by the theoretical equation of variation law. After analysis, the corrected 2f amplitude and temperature show a non-correlation and a correction for the effect of temperature on spectral detection is achieved. The shortcomings of the correction formula and the proposed improvement method are remedied, and the accuracy of the 2f amplitude correction at variable temperatures is verified. This study provides a reference for the practical application of spectral detection technology in the measurement process of high-temperature environments, especially for the dynamic evaluation of combustion efficiency in high-precision industrial furnaces, which is of great importance.