为扣除开放光路法红外光谱中的背景干扰 ,发挥开放光路法的优势 ,提出了针对多组分气体体系的小波变换(WT)和人工神经网络 (ANN)相结合的红外光谱定量方法。该方法在数据处理阶段利用小波变换方法扣除了样品光谱中的背景干扰 ,然后通过计算谱峰强度和组分浓度之间相关系数的方法确定了待测组分的特征峰 ,最后利用 ANN技术实现了定量分析。该方法用现场实测光谱进行了检验。结果显示其综合性能优于其它几种常用的方法 ,是一种有效的红外光谱定量方法 In order to deduce the background interference in the infrared spectrum of the open-light method and to take advantage of the open-path method, a quantitative infrared spectroscopy method based on wavelet transform (WT) and artificial neural network (ANN) for multicomponent gas system was proposed. In the data processing stage, the wavelet transform method is used to deduct the background interference in the sample spectrum. Then the characteristic peak of the component to be detected is determined by calculating the correlation coefficient between the peak intensity and the component concentration. Finally, the ANN technique is used Quantitative analysis. The method was tested with field measured spectra. The results show that its comprehensive performance is superior to several other commonly used methods, is an effective quantitative infrared spectroscopy