基于自适应多向独立成分分析的间歇过程监控的研究

Batch process monitoring based on adaptive multi-way independent component analysis

针对间歇过程批次与批次之间,操作条件缓慢变化的特性,提出一种基于自适应多向独立成分分析(MICA)的监控算法。该方法首先用MICA法建模,然后在历史数据集中加入新的正常批次并剔除最早批次,逐渐更新模型,同时引入遗忘因子,提高对新过程特性的适应性。青霉素发酵过程的仿真结果表明,自适应MICA比MICA更准确地描述过程行为,并有效减少检测故障时的误报。

Most industrial batch processes generally exhibit batch-to-batch variation in some degree. In this paper, an adaptive MICA method is proposed for batch process monitoring. This approach first gives an MICA model based on the historical database. The new batch data when monitored normally is added to the database and the oldest one is removed. On the basis of new database the old MICA model is revised by using forgetting factors to adapt to new normal conditions. The simulation results in monitoring fed-batch penicillin production show that the proposed approach effectively eliminates the false alarms generated by the fixed model.