论文部分内容阅读
针对间歇过程多阶段硬化分和误分类导致监控效果不理想的问题,提出了1种基于过渡时段分析的多阶段MPCA监控策略。该方法按照过程动态特性的变化,依据模糊C-均值算法(Fuzzy c-Mean,FCM)将过程数据划分为多个阶段,根据隶属度函数处理过渡阶段数据,建立阶段之间的联系;之后采用动态时间规整算法(Dynamic Time Warping,DTW)将分段后数据非线性化规整对齐,较好地解决了过渡阶段的监控问题,并通过在青霉素发酵过程的应用验证了该方法的有效性。
A multi-stage MPCA monitoring strategy based on transition period analysis is proposed to solve the problem that the monitoring results are not satisfactory due to the multi-stage hardening fraction and the misclassification of batch process. According to the change of the dynamic characteristics of the process, the method divides the process data into multiple stages according to the Fuzzy C-Mean (FCM) method and processes the transition stage data according to the membership function to establish the relationship between the stages. Then, The Dynamic Time Warping (DTW) algorithm uses the non-linearized alignment of the segmented data to solve the problem of monitoring in the transitional phase. The validity of this method is verified by the application in penicillin fermentation.