Fault detection method of MKPCA batch process based on Limited-DTW

Abstract

The invention discloses a fault detection method of an MKPCA batch process based on Limited-DTW. In order to overcome the inherent shortcomings of the batch unequal length of the intermittent process, and in order to overcome the serious defects of the traditional method of solving the inter-batch synchronization problem such as data waste, and the distortion of the variable autocorrelation and the cross-correlation relation between the variables of the original process, the method introduces the global path limit and the distortion threshold limit to improve the dynamic time warping (DTW) method, so as to avoid the monitoring shortcomings of the long-term operation of the algorithm and to solve the complexity of the processing and the practical application difficulty caused by the off-line performance. Limited-DTW and an MKPCA-based monitoring method are combined for offline and online implementation. The experiment design is completed by the penicillin fermentation simulation platform and the recombinant Escherichia coli practical production process, and the result displays the feasibility and validity of the method provided by the invention.

Description

technical field [0001] The invention relates to the technical field of data-driven multivariate statistical process monitoring (Multivariate Statistical Process Monitoring, MSPM), in particular to a Limited-DTW-based MKPCA intermittent process fault monitoring method. 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 the modern process industry, the batch process is widely used in the preparation of biomedicine, food and biochemical products, especially the preparation of penicillin (Penicillin, or transliterated penicillin). Compared with the continuous production process, the batch process has obvious differences, among which th...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problem of unequal length of batch data in batch process, the present invention provides a MKPCA batch process fault monitoring method based on Limited-DTW, which solves the problem of batch inconsistency by dynamically matching the original batch according to the pattern of midpoint and point in the trajectory. Isometric problem

Images