BDMOS:大数据挖掘优化系统及其在化工过程优化中的应用  

BDMOS: Big Data Mining Optimal System and its application in chemical process optimization

作　　者：徐潇[1] 赵洪涛 卢凯亮 张庆[4] 陆文聪[1,3] XU Xiao;ZHAO Hongtao;LU Kailiang;ZHANG Qing;LU Wencong(Department of Chemistry,College of Science,Shanghai University,Shanghai 200444,China;Shanghai 3F New Materials Co.Ltd,Shanghai 200241,China;Materials Genome Institute,Shanghai University,Shanghai 200444,China;Material Science and Engineering institute,Shanghai University,Shanghai 200444,China)

机构地区：[1]上海大学理学院,上海市200444 [2]上海三爱富新材料股份有限公司,上海市200241 [3]上海大学材料基因组工程研究院,上海市200444 [4]上海大学材料与科学工程学院,上海市200444

出　　处：《计算机与应用化学》2018年第6期433-439,共7页Computers and Applied Chemistry

基　　金：上海市科委重点项目“大数据挖掘和云计算技术在氟化工生产过程中的应用”(No.15dz1180700)

摘　　要：自主研发BDMOS（Big Data Mining Optimization System）大数据挖掘优化系统,该系统集成了多种数据挖掘方法和技术,特别是模式识别建模方法和优化技术。本工作将BDMOS大数据挖掘优化系统成功应用于聚全氟乙丙烯树脂反应装置及氟橡胶聚合生产装置的生产优化。根据装置DCS系统采集的生产数据,研究了装置优化操作的主要工艺参数,采用模式识别建模方法和优化技术建立了生产装置的优化操作模型,根据模型提出的工艺参数优化方案,显著提高了产品合格率。BDMOS软件提供的方法和技术能有效解决复杂化工数据的建模和优化难题,有望在化工生产装置优化过程中得到推广应用。The Big Data Mining Optimization System （BDMOS）, which independently developed, integrates multiple Data Mining methods and technologies, especially pattern recognition modeling and Optimization techniques. In this paper, BDMOS was successfully applied to the production optimization of polyfluorinated ethylene propylene resin reactor and fluorine rubber polymerization plant. According to the production data of DCS system, optimized operation parameters of the main technological device were studied. The equipment optimization operation model is established by pattern recognition and optimization technology, According to the optimization scheme of process parameters proposed by the model, the qualified rate of products was improved significantly. The method and technology provided by our BDMOS software can effectively solve the problems of modeling and optimization of complex chemical data, and it is expected to be popularized and applied in the process of chemical production plant optimization.

关 键 词：BDMOS 数据挖掘 模式识别 聚全氟乙丙烯树脂 氟橡胶 

分 类 号：TQ33[化学工程—橡胶工业] TQ322.3