A soft sensor modeling for aromatics yield based on adaptive weighted least squares support vector machine
The aromatics yield in catalytic reformate is an important quality indicator of the catalytic reforming process.Because of the difficulty in real-time measurement of the aromatics yield,building the soft sensor model for aromatics yield is particularly important.A novel adaptive weighted least squares support vector machine (AWLS-SVM) regression method is presented for the soft sensor modeling of aromatics yield.Firstly,in AWLS-SVM,least square support vector machine regression was employed for the sample data to develop model and obtain the sample datum fitting error.Secondly,the adaptive sample weights were obtained via the proposed improved exponential distribution weighted scheme according to the fitting error.Besides,the hybrid chaos differential evolution simulated annealing (CDESA) algorithm is proposed to obtain the optimal parameters of the LS-SVM.The simulation experiment results show that the outliers influence on the model's performance is eliminated in AWLS-SVM,and that the prediction performance is better than those of WLS-SVM and LS-SVM method.Furthermore,the AWLS-SVM is applied to develop the soft sensor model for aromatics yield,and the satisfactory result is obtained.
NETL
NSTL
万方数据
