CIESC Journal ›› 2012, Vol. 63 ›› Issue (12): 3978-3984.DOI: 10.3969/j.issn.0438-1157.2012.12.034

Previous Articles     Next Articles

Working condition-based optimization framework for operational patterns and its application in petrochemical industry

JIANG Baihua1, LIU Wei2, DAI Zhijun2, WANG Hong'an2   

  1. 1. Petro-Cyber Works Information Technology Co., Ltd., Beijing 100007, China;
    2. Institute of Software, Chinese Academy of Sciences, Beijing 100190, China
  • Received:2012-08-07 Revised:2012-08-17 Online:2012-08-29 Published:2012-12-05
  • Supported by:

    supported by the National Science and Technology Supporting Program(2012BAE05B03)and the National Basic Research Program of China(2009CB320804).

基于工况的操作模式优化及在石化工业中的应用

蒋白桦1, 刘伟2, 戴志军2, 王宏安2   

  1. 1. 石化盈科信息技术有限责任公司, 北京 100007;
    2. 中国科学院软件研究所, 北京 100190
  • 通讯作者: 刘伟
  • 作者简介:蒋白桦(1961-),男,硕士,高级工程师。
  • 基金资助:

    国家科技支撑计划项目(2012BAE05B03);国家重点基础研究发展计划项目(2009CB320804)。

Abstract: A new working condition-based optimization framework for operational patterns was proposed. The framework is composed of three components which are data preprocessing,the creation of library of optimized operational patterns and real-time optimization based on working conditions.On the basis of optimization and control framework,the application of online operational optimization for the dry-point temperature of FCCU was focused.Base on the FCCU properties,a two-step optimization approach for operational patterns which combines SVM and AdaBoost was proposed.Experimental results show that the new approach has higher prediction performance.

Key words: operational pattern, soft sensor, SVM, AdaBoost, pattern discovery

摘要: 提出了一种基于工况的操作模式优化框架,框架包括三部分:数据预处理、优化操作模式库的形成和基于工况的实时优化。在操作模式优化控制的框架和相关概念的基础上重点研究了催化过程中干点温度指标在线优化的应用。针对催化的工业特点,提出SVM与AdaBoost相结合的两步结合的操作模式的发现方法,试验证实该方法具有较高的预测性能。

关键词: 操作模式, 软测量, 支持向量机, AdaBoost, 模式发现

CLC Number: