A simulation model of the fiber-structure trace ammonia detection system based on near-infrared laser absorption spectroscopy is proposed. The distributed feed-back (DFB) diode laser modulation, photodetector, and signal processing are simulated by programming using matlab code while the noise amplitude selected to the sampling accuracy of commercial 12-bit ADC. The wavelength 1512 nm was selected to calculate the absorption coefficient of mixture containing ammonia, carbon dioxide, and water vapor. When the length of fiber-structure gas chamber was set to 10 cm, 1 m, and 10 m, respectively, ammonia with concentration from 0.2 ppm to 10,060 ppm can be detected. The dynamic range of this simulation have covered 5 orders of magnitude which never achieves in traditional detection system. Our work provides a potentially valuable option to establish the ammonia detection system of more portable, high sensitivity, and large detection range.

提出了一种基于近红外激光吸收光谱的纤维结构痕量氨检测系统的仿真模型。分布式反馈（DFB）二极管激光调制，光电检测器和信号处理是通过使用Matlab代码进行编程来模拟的，而噪声幅度则选择了商用12位ADC的采样精度。选择波长1512nm以计算包含氨，二氧化碳和水蒸气的混合物的吸收系数。当纤维结构气室的长度分别设为10cm，1m和10m时，可以检测到浓度为0.2ppm至10,060ppm的氨。该仿真的动态范围已覆盖了5个数量级，这是传统检测系统无法达到的。