基于RBF神经网络的生物质电站高温气体CO2浓度测量方法

电站气体浓度测量对实现燃烧优化、提高燃烧效率和火焰品质、减少污染物排放具有重要意义。以CO₂气体为例进行研究,基于近红外波段可调谐激光吸收层析成像技术,提出了基于径向基(radial basis function, RBF)神经网络的高温气体CO₂浓度测量方法。通过实验获取不同浓度下的CO₂吸收可调谐激光光谱信号,计算CO₂吸收谱线和原始信号的差值,提取出描述该差异性的统计特征参数作为RBF神经网络的输入,CO₂浓度作为RBF神经网络的输出,建立了基于RBF神经网络的高温气体CO₂浓度测量仿真模型,通过仿真实例验证了该方法的有效性和正确性。与GRNN神经网络对比分析表明:RBF神经网络法可以有效提高CO₂浓度测量精度,为生物质发电高温气体计量提供理论依据。

CO2 Concentration Measurement Method in High Temperature Gas of Biomass Power Plant Based on RBF Neural Network

The measurement of gas concentration in power plants is of great significance to realize combustion optimization,improve combustion efficiency and flame quality,and reduce pollutant emissions. Taking CO₂ as an example,according to infrared tunable laser absorption tomography technology,a measuring method of CO₂ concentration in high temperature gas based on radial basis function( RBF) neural network is proposed. The CO₂ absorption tunable laser spectral signals at different concentrations were obtained by experiments. The difference between the CO₂ absorption spectrum and the original signal is calculated,and the statistical characteristic parameters describing the difference are extracted,which are regarded as the input of RBF neural network and the CO₂ concentration as the output of RBF neural network. The model of the high temperature gas CO₂ concentration based on RBF neural network is established. The simulation results show that the method is effective and correct. Compared with GRNN neural network,the RBF neural network method can effectively improve the accuracy of CO₂ concentration measurement,and provide a theoretical basis for high temperature gas measurement in biomass power stations.