磨机报警智能化管理的分析与研究

Analysis and research on intelligent management of mill alarms

对照国际标准分析总结了磨机报警的种类和识别方法,并对国内磨机报警存在的问题进行了简要陈述,提出了利用密度函数和RCO曲线对磨机报警阈值优化的方法,以解决磨机报警阈值僵化的问题,利用Apriori算法对磨机报警数据库进行挖掘后对报警关联性进行分析,完成了基于定性分析的故障诊断和帮助系统的设计,同时对过程值变化率报警优化、报警优先等级的划分等方面进行了研究和分析,并提出了可行的智能化解决方案。

The paper summarized the types and identification methods of common mill alarms, and briefly described the problems existing in domestic mill alarms. After that, it proposed to apply the density function and RCO curve to optimize the alarm threshold, so as to solve the constancy of the mill alarm threshold. In addition, after Apriori algorithm being applied to dig the mill alarm database, the correlation among various alarms was analyzed, thus the fault diagnosis and the help system were designed based on qualitative analysis.Moreover, the optimization on the alarm for the process value variation rate and the ranking of alarm priority were studied, and then feasible intelligent solution was offered.