RBF neural network prediction of convention velocity in polymerizing process based on K-means clustering

• Published in: Proceedings of the 31st Chinese Control Conference pp. 3285 - 3290
• Main Authors: Wang Jiesheng, Zhu Jing, Guo Qiuping
• Format: Conference Proceeding
• Language: Chinese; English
• Published: IEEE 01-07-2012
• Subjects: Educational institutions; Electronic mail; K-means Clustering; Kernel; Kernel Principal Component Analysis; Polymerize Process; Polymers; Predictive models; Principal component analysis; Radial basis function networks; Radial Basis Function Neural Network
• ISBN: 1467325813; 9781467325813
• ISSN: 1934-1768; 2161-2927

For forecasting the key technology indicator convention velocity of vinyl chloride monomer (VCM) in the polyvinylchloride (PVC) polymerizing process, a predictive model based on radial basis function neural networks (RBFNN) is proposed. Firstly, kernel principal component analysis (KPCA) method is adopted to select the auxiliary variables of soft-sensing model in order to reduce the model dimensionality. Then the structure parameters of the RBFNN are optimized by the c K-means clustering method. In the end, simulation results show that the proposed model can significantly enhance the predictive accuracy and robustness of the technical-and-economic indexes and satisfy the real-time control requirements of PVC polymerizing production process.