基于匹配算法的脉冲差分吸收CO2激光雷达的稳频研究

利用差分吸收激光雷达探测大气CO₂, 可以获得其浓度的垂直分布, 对于研究碳源、碳汇的过程有重要意义. 设计了一套种子注入的脉冲差频激光器系统, 作为差分吸收激光雷达的激光光源. 针对脉冲差分吸收CO₂激光雷达on波长的高精度稳频的研究空白, 本文提出一种基于匹配的on波长的连续稳频算法. 其基本思想是采用分子饱和吸收法, 测量通过双路吸收池后的差分信号, 计算其光学厚度值(optical depth, OD), 获得实测的伪吸收谱, 当监测到on波长发漂移后, 进行连续的波长调节, 获取其OD值, 最后基于一维的图像匹配算法, 将OD值作为灰度值, 利用图像匹配原理, 进行OD值匹配, 确定当前输出波长在伪吸收谱中的位置, 进而调节至on波长, 实现on波长的连续、稳定输出. 实验结果表明, 提出的稳频算法能够很好的满足高精度的稳频要求, 同时差平方和法在该应用中是最优的, 稳频精度可达到0.3 pm.

Research on the frequency stabilization of pulsed differential absorbing lidar for CO2 detection based on matching algorithm

The differential absorption lidar (DIAL) can help us to obtain the vertical distribution of the atmospheric CO₂ concentration, which is important to the study of carbon sources and carbon sinks. We design a seeder injected pulsed laser system, working as the laser source of the CO₂ DIAL. Unlike the other CO₂ DIALs, our laser source is the result of difference frequency of two lasers at the wavelengths of 1064 nm and 634 nm, respectively. It should be pointed out that the high frequency (wavelength) accuracy and stability of the emission laser, especially the on-line one, are greatly required in the CO₂ DIAL system. However, the mechanical properties of the dye laser (634 nm) and the application of laser difference frequency technique make the wavelength drift constantly; besides, the extremely unstable energy of the pulsed laser increases the difficulty in identifying and stabilizing the on-line wavelength. Hence, a fast and efficient frequency (wavelength) stabilization method is needed to achieve a stable emission wavelength. Aiming at the research gap of the high precision requirements of on-line laser for this kind of pulsed DIAL, we propose a frequency stabilization method based on matching algorithm. The basic idea is to utilize the saturable absorption of CO₂ molecule, by measuring the differential residual-intensity after the laser passing through dual absorption cells to calculate the optical depth (OD) and obtain the so-called pseudo CO₂ absorption spectrum, which can be used to identify the on-line laser accurately. Finally, based on the matching algorithm of one-dimensional image, treating the OD as the gray value in the image, we implement the OD matching as a most important part in the process of frequency stabilization, and determine the exact position of the real-time output laser in the measured pseudo absorption spectrum. Thus, when some errors happen to the monitored ODs, by continuously adjusting the wavelength of the laser, the proposed method can fulfill the wavelength adjustment and accomplish the continuous frequency stabilization for on-line laser. Experimental results show that the frequency stabilization algorithm based on OD matching can satisfy the requirements for pulsed on-line laser frequency stabilization, and the sum of squares of deviation method is the optimal one in this application, with a stabilization accuracy of 0.3 pm. Besides, the proposed method can also be introduced in other laser frequency stabilization.