热力发电2017，Vol.46Issue(4)：81-87,7.DOI:10.3969/j.issn.1002-3364.2017.04.081
基于偏最小二乘支持向量机的烟气湿法脱硫效率预测模型
Prediction model for flue gas wet desulfurization efficiency based on partial least squares support vector machine
摘要
Abstract
In order to better reflect the relationship between the wet flue gas desulfurization process parameters and the desulfurization efficiency,the partial least squares regression (PLS) method was applied to analyze the process factors affecting the flue gas desulfurization efficiency,then the factors affecting the desulfurization efficiency greatly were considered as the principal components and were extracted for prediction using the support vector machine (SVM).Thus,on the basis of the partial least squares support vector machine (PLS-SVM),a prediction model for wet flue gas desulfurization efficiency was established,which can reduce the input dimensions of the SVM.Moreover,the data of the operation monitoring system of the limestone gypsum wet desulphurization facility were selected to carry out the training and prediction for the model.The prediction results show that,the maximum absolute prediction error of the PLS-SVM model is below 0.65％,and the average absolute error is about 0.3％,indicating the above model can well predict the desulfurization efficiency,and its prediction efficiency and accuracy is higher than that of the SVM model.In order to better reflect the relationship between the wet flue gas desulfurization process parameters and the desulfurization efficiency,the partial least squares regression (PLS) method was applied to analyze the process factors affecting the flue gas desulfurization efficiency,then the factors affecting the desulfurization efficiency greatly were considered as the principal components and were extracted for prediction using the support vector machine (SVM).Thus,on the basis of the partial least squares support vector machine (PLS-SVM),a prediction model for wet flue gas desulfurization efficiency was established,which can reduce the input dimensions of the SVM.Moreover,the data of the operation monitoring system of the limestone gypsum wet desulphurization facility were selected to carry out the training and prediction for the model.The prediction results show that,the maximum absolute prediction error of the PLS-SVM model is below 0.65％,and the average absolute error is about 0.3％,indicating the above model can well predict the desulfurization efficiency,and its prediction efficiency and accuracy is higher than that of the SVM model.关键词
偏最小二乘回归/支持向量机/湿法脱硫/脱硫效率/预测模型Key words
partial least squares regression/support vector machine/wet desulfurization/desulfurization efficiency/prediction model分类
资源环境引用本文复制引用
崔仕文,铁治欣,丁成富,赵峰..基于偏最小二乘支持向量机的烟气湿法脱硫效率预测模型[J].热力发电,2017,46(4):81-87,7.基金项目
浙江省公益技术应用研究项目(2014C31G2060072)Public Feehnology Application Projects of Zhejiang Province (2014C31G2060072) （2014C31G2060072）
