基于自组织映射模型的数据挖掘算法在对二甲苯吸附分离过程中的应用

Application of SOM-based Data Mining Methods to PX Absorption and Separation Process

  • 摘要: 以PX吸附分离过程为研究对象,运用基于SOM模型的数据挖掘算法对其进行分析研究.SOM模型在整个挖掘过程中起了关键性的作用.一方面,SOM模型作为探索性数据分析的有效工具,为进一步的挖掘提供了依据.另一方面,SOM模型为聚类算法提供参数指导和数据支持.最终,通过数据挖掘实现了两个目标,得到了在不同负荷情况下操作参数的稳态优化区域;建立了可用于指导操作员改进操作的可视化实时评估模型.

     

    Abstract: PX absorption and separation process is studied and analyzed using data mining methods based on SOM model,and the SOM model plays a key role in the whole data mining process.On the one hand,SOM model is used as an effective exploratory data analysis tool and provides evidence for further data mining.On the other hand,SOM algorithm provides directions in parameter selection and offer data support for other data mining algorithms(such as clustering,rule extraction,etc).In conclusion,two aims are obtained with the data mining algorithms: the optimized zone of operating parameters in different loads is obtained,and a realtime visualized evaluation model is built to improve the operation performances.

     

/

返回文章
返回