Development of Adversarial Transfer Learning Soft Sensor for Multigrade Processes

Liu, Yi; Yang, Chao; Zhang, Mingtao; Dai, Yun; Yao, Yuan

doi:10.1021/acs.iecr.0c02398

Cited by 66 publications

(39 citation statements)

References 38 publications

(85 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Liu et al (2019) combined a domain adaptation extreme learning machine (DAELM) and transfer learning to establish a simple soft sensor model suitable for multi‐grade processes with limited labeled data. Based on the DAELM model, Liu et al (2020) introduced an adversarial transfer learning to learn a suitable feature transformation between different domains and improve the performance of model. Yuan et al (2020) implemented soft sensor models based on a new multichannel convolutional neural network to extract the local dynamic and nonlinear features of process data for accurate prediction.…”

Section: Introductionmentioning

confidence: 99%

Soft sensor with deep feature extraction for a sugarcane milling system

Meng

Chen

et al. 2022

J Food Process Engineering

View full text Add to dashboard Cite

Extraction rate and energy consumption are two important indicators of the sugarcane milling system, which are calculated by offline experiments and inspection methods. Soft sensors are inferential models, which are enabling online prediction of these indicators. Selection of input variables is one of the most critical issues in soft sensors design. So a deep feature extraction method is proposed by combining mutual information theory and hybrid chicken swamp algorithm to determine the input and parameters of model. Then, input the obtained variables into the data‐driven model of deep kernal extreme learning machine (DK‐ELM) with optimized parameters to realize the prediction of extraction and hourly energy consumption. It combines the kernel method with the multi‐layer ELM. Experiments are conducted to verify the effectiveness and superiority of the modeling method. Comparing the proposed framework with the four experiments designed on a sugarcane milling system, the proposed framework for predicting extraction and energy consumption with better feasibility and effectiveness, and provided a good reference for end‐point control and judgment of quick direct tapping. Practical Applications Extraction and energy consumption are two important indicators of the milling process and whether they meet standards will affect the smooth operation and economic benefits of sugar production. This study presents a novel modeling method to predict these indicators based on the deep feature extraction and deep kernal extreme learning machine. These methods provided a good reference for end‐point control and judgment of quick direct tapping. In addition, it a new concept for the design of soft sensors for such complex industrial processes.

show abstract

Section: Introductionmentioning

confidence: 99%

Soft sensor with deep feature extraction for a sugarcane milling system

Meng

Chen

et al. 2022

J Food Process Engineering

View full text Add to dashboard Cite

show abstract

“…A few applications to SS design have been proposed in very recent works. In Reference [ 64 ], a domain adaptation, soft sensing framework for multi-grade chemical processes is discussed. A limited number of labeled samples is available for some operating grades.…”

Section: Related Workmentioning

confidence: 99%

RNN- and LSTM-Based Soft Sensors Transferability for an Industrial Process

Curreri

Patané

Xibilia

2021

Sensors

View full text Add to dashboard Cite

The design and application of Soft Sensors (SSs) in the process industry is a growing research field, which needs to mediate problems of model accuracy with data availability and computational complexity. Black-box machine learning (ML) methods are often used as an efficient tool to implement SSs. Many efforts are, however, required to properly select input variables, model class, model order and the needed hyperparameters. The aim of this work was to investigate the possibility to transfer the knowledge acquired in the design of a SS for a given process to a similar one. This has been approached as a transfer learning problem from a source to a target domain. The implementation of a transfer learning procedure allows to considerably reduce the computational time dedicated to the SS design procedure, leaving out many of the required phases. Two transfer learning methods have been proposed, evaluating their suitability to design SSs based on nonlinear dynamical models. Recurrent neural structures have been used to implement the SSs. In detail, recurrent neural networks and long short-term memory architectures have been compared in regard to their transferability. An industrial case of study has been considered, to evaluate the performance of the proposed procedures and the best compromise between SS performance and computational effort in transferring the model. The problem of labeled data scarcity in the target domain has been also discussed. The obtained results demonstrate the suitability of the proposed transfer learning methods in the design of nonlinear dynamical models for industrial systems.

show abstract

“…Step 7. Given the values of parameters α and σ 2 and the training set fs tr , y tr g, compute the posterior mean μ and covariance Σ of the model according to equations (26) and (27).…”

Section: Journal Of Sensorsmentioning

confidence: 99%

“…To this end, nonlinear modeling methods are used in MI prediction, such as artificial neural networks (ANNs) [17], support vector machines (SVMs) [18], Gaussian process regression (GPR) [19,20], and relevance vector machine (RVM) [21] [ [22][23][24][25]. Recently, Liu et al proposed an adversarial transfer learning-(ATL-) based soft sensor [26] and a domain adaptation transfer learning soft sensor for product quality prediction [7]. As classical methods, PCA and PLS have achieved great successes with respect to quality prediction by extracting latent variables (LVs) [27][28][29].…”

Section: Introductionmentioning

confidence: 99%

Soft Sensor Development Based on Quality‐Relevant Slow Feature Analysis and Bayesian Regression with Application to Propylene Polymerization

et al. 2021

View full text Add to dashboard Cite

Data-driven soft sensors are widely used to predict quality indices in propylene polymerization processes to improve the availability of measurements and efficiency. To deal with the nonlinearity and dynamics in propylene polymerization processes, a novel soft sensor based on quality-relevant slow feature analysis and Bayesian regression is proposed in this paper. The proposed method can handle the dynamics of the process better by extracting quality-relevant slow features, which present both the slowly varying characteristic and the correlations with quality indices. Meanwhile, a Bayesian inference model is developed to predict the quality indices, which takes advantages of a probability framework with iterative maximum likelihood techniques for parameter estimation and a sparse constraint for avoiding overfitting. Finally, a case study is conducted with data sampled from a practical industrial propylene polymerization process to demonstrate the effectiveness and superiority of the proposed method.

show abstract

Development of Adversarial Transfer Learning Soft Sensor for Multigrade Processes

Cited by 66 publications

References 38 publications

Soft sensor with deep feature extraction for a sugarcane milling system

Soft sensor with deep feature extraction for a sugarcane milling system

RNN- and LSTM-Based Soft Sensors Transferability for an Industrial Process

Soft Sensor Development Based on Quality‐Relevant Slow Feature Analysis and Bayesian Regression with Application to Propylene Polymerization

Contact Info

Product

Resources

About