主元分析及多变量过程监测联合预测N Ox 质量浓度
赵志宏;韩超;赵文杰
【期刊名称】《热力发电》 【年(卷),期】2016(045)007
【摘要】针对火电厂烟气N Ox 质量浓度检测分析仪存在延时较长的问题，本文首先采用主元分析及多变量过程监测选出影响脱硝反应器入口烟气 N Ox 质量浓度的主导因素，再根据分析仪的大致延时时间Ts ，选取Ts 后反应器入口烟气N Ox 质量浓度数据为训练信号，进而利用最小二乘支持向量机(LS-SVM)建立反应器入口烟气 NOx 质量浓度的预测模型。采用某电厂历史运行数据对该预测模型进行训练及测试，结果表明：利用主元分析及多变量过程监测方法从多个辅助变量中选出主导因素，简化模型结构的同时也使预测模型具有较好的学习及泛化能力。该方法是反应器入口烟气N Ox 质量浓度准确、实时预测的一种可行方法。?cording to serious decay of NOx emission analyzer used in thermal power plants,the principal component analysis and multivariable process monitoring method were employed to select the dominant factors influencing the NOx concentration at entrance of the denitration reactor.Then,on the basis of the analyzer's decay time,the NOx content data of the flue gas at inlet of the reactor was selected as the peda-gogic signal to establish the prediction model for NOx concentration in flue gas at inlet of the reactor,by u-sing the least squares support vector machine (LS-SVM).Moreover,combing with the historical operation data of a power plant,the above model was trained
主元分析及多变量过程监测联合预测N Ox 质量浓度
