Abstract: The database of real-time control system accumulates a great number of system data. These data have concealed with many operator control methods that related to the status tendency of industrial process. And they have laid the foundation for the application of data mining technology in the field of industrial process intellectual control. But mining the expert rule in the continuously numerical data that needs not to have the fast valid method at present. In this paper, the author propose a method which translate numerical type data into character type date and then mine the frequent sequence that in order to obtain the rule of fuzzy expert control system, and achieved the intelligence control of the gas mixture pressure station in the iron and steel factory .

Key words: Indirect Inter-Relevant Successive Tress with Dual Arties, fuzzy expert system, acquirement of expert knowledge

摘要： 在系统的实时数据库中，积累了大量的系统运行数据，其中隐含了与工业过程运行状态趋势相关的操作人员控制方法，这为数据挖掘技术在工业过程智能控制中的应用奠定了基础。但是要在连续数值型的数据中挖掘出所需要的专家规则目前还没有快速有效的方法。本文提出了将数值型数据转换成字符型数据再进行频繁序列挖掘的方法，以获取模糊专家系统的控制规则，并在钢铁厂实现了煤气混合加压站智能控制。

关键词: 二元间接互关联后继树、模糊专家系统、专家知识获取