Information transmission based phase affiliation judgment method for real time sampling points in intermittent process

Abstract

An information transmission based phase affiliation judgment method for real time sampling points in an intermittent process relates to the technical field of data driving based on multivariate statistics process monitoring (MSPM). According to the invention, phase affiliation judgment of the real time sampling points subjected to online monitoring is realized on the basis of failure detection in a multi-phase intermittent process and the monitoring performance is improved. In order to guarantee the safe and stable operation of a penicillin fermentation process and also to improve the effectiveness of a prior penicillin fermentation process monitoring method, segmentation modeling on batch data of the fermentation process is an effective means for improving model precision. According to the invention, aiming at a problem of how to choose a monitoring model of a corresponding phase accurately for the real time monitoring sampling points during online monitoring by the segmentation modeling in the penicillin fermentation process mainly and through calculating message scales of the clustering centers obtained through calculating online real time sampling points and offline phase division, stable phase affiliation results are output stably through information iteration and model selection during online monitoring can be guided.

Description

technical field [0001] The invention relates to the technical field of data-driven multivariate statistical process monitoring (MSPM), in particular to a method for judging the stage attribution of real-time sampling points during online monitoring of multi-stage intermittent processes. The present invention realizes stage attribution judgment of real-time sampling points during on-line monitoring on the basis of multi-stage intermittent process fault detection, and improves monitoring performance. Background technique [0002] When the data-driven multivariate statistical method is used for the process monitoring of the batch process, it does not need to consider the complex process mechanism characteristics. Through the modeling and analysis of historical data, it can be judged whether the operation status of the production process is abnormal. As one of the important production methods in modern process industry, batch process is widely used in the preparation of biomedic...

AI Technical Summary

Problems solved by technology

However, the above methods are all based on a default assumption that the data at the same moment belong to the same stage, that is, the same stage is of equal length in all production batches

However, this is not the case in the actual production process. The production of different batches is different due to various conditions, resulting in differences in the length of the internal stages of the process between different batches, such as figure 1 shown

Therefore, the stage assignment of real-time sampling points based on time may divide the data of different stages into one category, thereby affecting the accuracy of the monitoring model, resulting in a large number of false alarms and missed alarms

Images