Non-linear process monitoring method

• Main Authors: XIAO YINGWANG, CHEN CHENGGUO, LIU JUN, ZHANG XUHONG, LI LI, YAO MEIYIN
• Format: Patent
• Language: Chinese; English
• Published: 15.09.2017
• Subjects: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION ORPROCESSING OF GOODS; CONTROL OR REGULATING SYSTEMS IN GENERAL; CONTROLLING; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS ORELEMENTS; PHYSICS; REGULATING; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS; TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINSTCLIMATE CHANGE

The present invention discloses a non-linear process monitoring method used for monitoring of an industrial production process. The non-linear process monitoring method employs the idea of the Fisher discrimination function to construct a target function for optimizing parameters of a kernel function, employs the particle swarm optimization algorithm to search the optimal solution of the target function as the optimal kernel parameters of the kernel function, and employs the method of kernel function principal component analysis to determine the reasons causing the process monitoring faults so as to solve the processing monitoring non-linear data processing, reduce the dimension of the process monitoring variable and improve the process monitoring efficiency and accuracy. 本发明公开了种非线性过程监控方法，用于工业生产过程监控。本发明采用Fisher判别函数的思想构建优化核函数参数的目标函数，利用粒子群优化算法寻求目标函数最优解作为核函数的最优核参数，采用核函数主元分析的方法确定导致过程监控故障的原因，解决过程监控非线性的数据处理，降低过程监控变量的维度，提高过程监控的效率和准确度。