Soft Sensor Modeling Based on PCA and Support Vector Machines

Soft sensor is an effective method to estimate variables which are difficult to be measured on-line in industrial processes.Support vector machine(SVM)is a novel machine learning method based on the statistical learning theory.A soft sensor based on principal component analysis(PCA)and Least Square SVM was proposed.The PCA method could not only solve the linear correlation of the input and compress data but also simply the SVM structure.Cross validation method was used to select parametrs of LS-SVM model.Soft sensor was applied to prediction of 4-CBA.Results indicates that this method features high learning speed,good approximation and well generalization ability.It provides convenice for on-line 4-CBA measurement.