Complex working condition measurement method based on improved GWO optimized SVM

• Main Authors: QIN WEI, CONG SHUOFENG, ZHANG XIBIN, LUO WANBO, LI HUI, ZHANG XIUMEI, SUN LEI, ZHAO QILIANG, CHEN HAO, LIU YUE, JIANG ZHIYU, LUO MINGYUE, CHU ZHUANG
• Format: Patent
• Language: Chinese; English
• Published: 18.10.2022
• Subjects: ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVEREDIN A SINGLE OTHER SUBCLASS; CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; MEASURING; MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS; TARIFF METERING APPARATUS; TESTING

The invention belongs to the field of industrial process intelligent control, and particularly relates to a complex working condition measurement method based on an improved GWO optimized SVM. According to the main content of the method, auxiliary variables are selected according to a specific extrusion process mechanism and on-site working conditions through priori knowledge, a soft measurement model of the polymer extrusion process is established based on a support vector machine (SVM) through data preprocessing, and supervised learning is performed on the soft measurement model. A penalty coefficient C and a kernel function coefficient G of an SVM (Support Vector Machine) are optimized by adopting a Grey Wolf Optimization (GWO) algorithm improved by a reverse learning strategy, and a soft measurement model is verified, so that the problems of difficult acquisition of system process parameters, difficult establishment of a prediction model, relatively large lag in a control process and the like in a polymer