节点文献

基于数据驱动的催化裂化过程反应-再生单元早期故障检测(英文)

A data-driven approach for early fault detection of the reactor regenerator in the RFCC unit

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 KIALAIN Kaustella杨洋李颖张新宇杜增智IFTIKHAR Maria杨彦宇孙巍

【Author】 KIALAIN Kaustella;YANG Yang;LI Ying;ZHANG Xinyu;DU Zengzhi;IFTIKHAR Maria;YANG Yanyu;SUN Wei;Key Lab of Membrane Science and Technology, College of Chemical Engineering, Beijing University of Chemical Technology;Shunhua College of Energy, Liaoning University of Petro-chemical Technology;Center for Process Simulation and Optimization, Beijing University of Chemical Technology;

【机构】 北京化工大学化学工程学院膜分离过程与技术北京市重点实验室顺华能源学院辽宁石油化工大学北京化工大学过程模拟优化中心

【摘要】 反应-再生装置是渣油催化裂化的核心部分。反应器的性能改变会直接影响产品的质量。在一个工业过程中,一般的故障都是自然或者突然产生的。大量的研究论文都致力于监测的突然故障,但发现早期故障也很重要,因为它们特点是缓慢的动态变化,最终会导致更严重的问题。建议使用的方法称为CUSUM-PCA,与传统的PCA方法相比较具有早期故障检测功能以及高级过程监控的特点。结果表明,使用CUSUM-PCA方法比传统的PCA方法提前一个小时检测出异常。

【Abstract】 The reactor–regenerator is the heart of the RFCC unit. Change in the performance of the reactor directly influences product quality. Usually faults in an industrial process can either be abrupt or incipient in nature. Significant number of research papers have been dedicated to the monitoring of abrupt faults but uncovering incipient faults is also important, as they are characterized by their slow dynamic change that eventually leads to a more serious problem. The use of a proposed method known as CUSUM-PCA is compared with conventional PCA for its early fault detection features and advanced process monitoring characteristics. Results show that the proposed CUSUM-PCA provides abnormal situation detection over an hour in advanced than conventional PCA.

【基金】 Fundamental Research Funds for the Central Universities(ZY1511)
  • 【文献出处】 计算机与应用化学 ,Computers and Applied Chemistry , 编辑部邮箱 ,2015年11期
  • 【分类号】TE624.41
  • 【下载频次】98
节点文献中: 

本文链接的文献网络图示:

本文的引文网络