基于遗传算法的小波神经网络在多组分气体检测中的应用

由于利用不分光红外吸收法（NDIR）的多组分气体传感器对汽车尾气（主要成分为CO2、CO、HC化合物）进行同时测量时，所测气体浓度是交叉吸收干扰后的结果，造成测量误差大，分析精度低。针对此问题，将遗传算法优化的小波神经网络用于建立基于红外光谱的三组分气体定量分析模型中。采集CO2、CO、HC的浓度信号，作为模型输入，通过模型回归分析，得到对应的混合气体组分浓度，解决气体之间相互干扰的问题。最后通过实验数据对模型性能进行仿真分析，结果表明，该模型的平均误差相比于传统模型明显减低，取得较好的精度。

Application of wavelet neural network based on Genetic Algorithm in multi component gas detection

Because of the simultaneous measurement of automobile exhaust gas by using the multi-component gas sensor,the gas concentration is the result of the cross absorption and interference,resulting in the large measure?ment error and low accuracy. Aiming at this problem,the genetic algorithm optimization of the wavelet neural net?work is used to establish the three component gas quantitative analysis model based on infrared spectrum. The con?centration signals of CO2,CO,HC,as the model input,through the model regression analysis,to get the correspond?ing mixed gas concentration and solve the problem of mutual interference. Finally,the model performance is simu?lated by the experimental data. The results show that the average error of the model is significantly reduced com?pared to the traditional model.