Fault Diagnosis of Cta Hydrogenation Process Based on KPCA

• 作 者: BoLi;WeiminZhong;ZhiLi;NaLuo;FengQian;
• 机 构: East China University of Science and Technology,China
• 出 处: 第三届先进计算智能与智能信息学国际研讨会(The Third International Workshop on Adv
• 发表日期: 2013年期

　　Because the large amount of operating data collected from Distributed Control System(DCS)often contain outliers,and the data are more complexity,nonlinearility.They cant be used directly to model,optimization and fault diagnosis.In fault diagnosis,the existence of outliers can destroy the covariance structure of Kernel Principal Component Analysis(KPCA),which leads to the model cant really reflect the actual normal conditions.In this paper,KPCA method is adopted to establish the normal statistical monitoring model from the historical data which can represent the normal industrial operating conditions.First,the outliers detection algorithm is used to eliminate outliers among normal working condition data,then the primary statistical model for fault diagnosis of the Squared Prediction Error(SPE)and T2 is established according to the data excluding outliers.The effectiveness of this fault diagnosis is demonstrated by the operating data of industrial Crude Terephthalic Acid(CTA)hydrogenation process,and simulation results show that this method can identified the industrial failure conditions.