Improved Mahalanobis Distance Based JITL-LSTM Soft Sensor for Multiphase Batch Processes

Feifan Shen,Jiaqi Zheng,Lingjian Ye

doi:10.1109/access.2021.3079184

Feifan Shen, Jiaqi Zheng + Show 1 more

Open Access

https://doi.org/10.1109/access.2021.3079184

Copy DOI

Abstract

To predict key variables of complicated batch processes, the long short-term memory (LSTM) soft sensor is developed to deal with both data nonlinearity and dynamics. To extract proper historical samples and implement the real-time modeling scheme with model updating strategy, the just-in-time learning (JITL) algorithm is widely used at the data selection stage of LSTM soft sensor. However, the multiphase issue of batch processes are not considered for the conventional JITL-LSTM soft sensor. In this paper, a multiphase Mahalanobis distance based JITL framework is developed to integrate the phase recognition strategy into the similarity measurement and data selection scheme, by which an extra step of phase identification can be avoided and the accuracy of JITL can be significantly improved. Thus, batch samples from different operating phases can be recognized without an additional phase identification step. By the use of the Mahalanobis Distance based JITL-LSTM Soft Sensor, the probability of data mismatch can be significantly reduced so that the accuracy of quality prediction can be promoted. Two simulation cases are provided to verify the effectiveness of the proposed method consisting of a fed-batch reactor process and the penicillin fermentation process.

Highlights

Both batch and continuous processes are important modes of production in modern industry
The results definitely demonstrate the advantages of the proposed method in dealing with the quality prediction problem of multiphase batch processes
In this work, a multiphase Mahalanobis distance based JITL (MMJITL)-long short-term memory (LSTM) soft sensor framework is developed for quality prediction of multiphase batch processes

Summary

Introduction

Both batch and continuous processes are important modes of production in modern industry. With the rapid development of process industry, the proportion of batch processes is growing due to the increasing demand of highvalue products such as pharmaceuticals, polymers and semiconductors [1]–[6]. In terms of the situation, researchers focus on the quality prediction and monitoring problem of batch processes to ensure the product quality and process safety. The related research starts late so that it is common to apply the soft sensors designing for continuous processes to batch processes. Different from the continuous processes with steady operating condition, the batch process. The statistical characteristics of the data collected from the batch process is more complicated than the continuous process [7]–[9]

Objectives

Results

Conclusion