基于时间序列分析与数据模型的过程报警管理系统

发布时间：2018-05-17 17:36

本文选题：时间序列分析 + 预测建模　；参考：《电子科技大学》2014年硕士论文

【摘要】：流程工业生产通常在高温、高压等极端恶劣的环境下作业,事故的发生几率较大,后果严重。在现代工厂中,配置在大型复杂工业过程中的集散控制系统以及紧急停车系统,大大提高了自动化控制水平,也使过程工业中大量有价值的信息蕴含其中,这之中也包含了报警信息。但大部分报警系统只能获取表层的报警信息,并不能分析、处理这些数据,使其充分发挥价值。如果操作人员能利用报警系统对工业测控系统中参数变化情况进行监测,在报警系统的参数发生异常时发出预警,操作人员会做出及时的处理;或者即使无法及时发现参数变化,也会因报警系统发出报警,使得事故得以避免,从而减轻操作人员心理压力并缓解工作强度。集散控制系统数据库中记录了大量的历史数据,这些数据随着过程安全问题的重视,在流程工业中得到更多利用。因而,利用工业参数变化来进行工业现场预警有着非常重要的意义。本文首先介绍了时间序列分析方法的研究进展,并介绍了时间序列分析方法的基本概念以及建模过程,提出了一种过程报警的预测方法,使用VB技术、SQL数据库技术以及OPC技术设计了过程报警管理系统,对OPC服务器以及客户端的程序进行了设计。将时间序列模型与过程报警状态序列和持续时间相结合,通过Matlab的TE平台实现对现场监测数据的处理以及建模过程,并在VB中使用ActiveX技术实现对Matlab处理结果的调用,实现了对我公司现场数据的监测、预警以及报表统计等功能。
[Abstract]:The process industrial production usually works under the extremely bad environment such as high temperature and high pressure, the accident occurrence probability is big, the consequence is serious. In modern factories, distributed control systems and emergency parking systems, which are deployed in large-scale and complex industrial processes, greatly improve the level of automation control, and make a large amount of valuable information contained in the process industry. It also contains an alarm message. However, most alarm systems can only obtain the alarm information in the surface layer, and can not analyze and process these data, so that they can give full play to the value. If the operator can use the alarm system to monitor the change of the parameters in the industrial measurement and control system, and issue an early warning when the parameters of the alarm system are abnormal, the operator will make timely treatment; Or even if the change of parameters can not be detected in time, the alarm system will send out an alarm, so that the accident can be avoided, thereby reducing the psychological pressure of the operator and alleviating the working intensity. A large number of historical data have been recorded in the database of distributed control system. With the attention of process security, these data have been used more and more in the process industry. Therefore, it is very important to make use of the industrial parameter change to carry on the industrial spot early warning. This paper first introduces the research progress of time series analysis method, introduces the basic concept of time series analysis method and the modeling process, and puts forward a forecasting method of process alarm. The process alarm management system is designed by using VB technology, SQL database technology and OPC technology, and the program of OPC server and client is designed. This paper combines the time series model with the process alarm state sequence and duration, realizes the processing and modeling process of the field monitoring data through the te platform of Matlab, and uses the ActiveX technology in VB to realize the call to the Matlab processing result. Realized the monitoring of our company's site data, early warning and report statistics and other functions.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TB496;TP277

【参考文献】