基于支持向量机的曝气池SVI软测量研究

支持向量机(Support Vector Machines)是一种基于统计学习理论的新型建模方法。活性污泥法工艺中反应出水水质的重要参数(曝气池SVI)难于在线测量,为此用某城市污水处理厂的实际水质参数数据,采用粗糙集理论(RS)和主元分析(PCA)进行数据的预处理,建立了基于SVM的曝气池SVI软测量模型,并进行了比较。综合考虑后,选择RS-PCA-SVM模型。结果表明,经粗糙集和主元分析预处理数据后,大大降低了输入数据的维数,减小了模型的规模,更有利于软测量模型的实用化。同时也表明支持向量机作为建立软测量模型的工具,具有良好的性能。

Studying About Aeration Tank SVI Soft Measure Based On SVM

Support Vector Machines is a kind of new machine learning method based statistic.In this paper,the aeration tank SVI soft measure model based on SVM established with RS and PCA,using practical data of the water quality parameters of some wastewater treatment plant.After contact,choice RS-PCA-SVM.The result indicate that the dimensions of the input data are reduced greatly when RS is used,and SVM is advantage as method establishing soft measure model than artificial neural network.