LSTM Soft Sensor Development of Batch Processes With Multivariate Trajectory-Based Ensemble Just-in-Time Learning

Abstract

To implement the quality prediction scheme for batch processes, long short-term memory (LSTM) neural network is a feasible tool to handle with the process dynamics and nonlinearity. However, a global LSTM soft sensor suffers a decline in performance facing batch-to-batch variations. To overcome the batch diversity problem and take advantage of LSTM model, a multivariate trajectory based ensemble just-in-time learning strategy is proposed in this paper. Different trajectory based similarity measurements are designed to extract historical batch trajectories which share similar spatial positions and trends. For each selected trajectory, an online local LSTM soft sensing model is constructed and the real-time quality prediction result for each local model can be obtained. Then, a weighting parameter is determined for each model by cross validation. Bringing together quality prediction results from different local models, the ensemble prediction result can be finally figured out. Two case studies are carried out to prove the effectiveness of the proposed methodology including a fed-batch reactor and the fed-batch penicillin fermentation process.INDEX TERMS Batch production systems, Ensemble just-in-time learning, long short-term memory, multivariate trajectory analysis, soft sensor, quality prediction.

Ensemble Learning Applications in Multiple Industries: A Review

“…In the food material industry, 3Dprint technology is used and random forest (RF) is integrated to improve the performance. In the food processing industry, RF and gradient boosting regression (GBR) algorithm is used to optimize pellet quality [43], in addition to the abovementioned fields, In paper industry [44], chemical industry [45], pharmaceutical manufacturing [46], plastic products [47], ferrous metal [48], non-ferrous metal [49], automobile manufacturing industry [50], transportation industry [51], through method based on EL innovation, data innovation, domain innovation to enhance and improve the existing manufacturing process has obtained the best benefits. Table 4 shows application of EL has grown by leaps and bounds.…”

A semi‐supervised JITL paradigm based on manifold regularization for online soft sensor development

“…For example, Shen et al applied similarity metrics based on spatial, angle, and batch trajectories in integrating JITL to cope with the batch diversity problem. 11 Yuan et al developed a spatial-temporal adaptive LWPLS model for JITL by considering spatial and temporal similarity. 12 Nevertheless, the above JITL methods usually require sufficient quantity of labeled samples.…”

Nonlinear Dynamic Soft Sensor Development with a Supervised Hybrid CNN-LSTM Network for Industrial Processes

“…The long-term memory is taken into consideration for LSTM, which is able to describe the time-series model more accurately with more parameters in comparison to RNN. So far, LSTM-based soft sensors have been successfully designed and applied to different industrial processes with both nonlinear and dynamic properties. , However, soft sensor models based on the conventional RNN and LSTM structures are unsupervised, which means that the quality information may not been exploited in the hidden units. To make full use of the quality data, a soft sensor model based on a dynamic neural network named nonlinear autoregression with exogenous input (NARX) was designed .…”