In industrial processes, the change in working conditions contributes to the distribution divergence between training data and test data. However, conventional soft sensor approaches follow the assumption that training data and test data follow the same data distribution. Consequently, the established soft sensor model is inapplicable to new working conditions. In order to tackle this challenge, a novel approach based on discriminative and statistical transfer learning (DSTL) is proposed for soft sensors under multiple working conditions. First, the -means clustering is applied to establish class pseudo labels. Besides, statistical distribution alignment is introduced to overcome marginal and conditional data distribution discrepancies between historical working conditions and new working conditions. Finally, by minimizing intra-class discrepancy and maximizing inter-class discrepancy, DSTL preserves discriminative information across multiple working conditions, which in turn enhances the performance of statistical distribution alignment. Comprehensive experiments on a fed-batch penicillin fermentation process are adopted to verify the effectiveness of the proposed model.
AlakentB (2021) Soft-sensor design via task transferred just-in-time learning coupled transductive moving window learner. Journal of Process Control101: 52–67.
2.
BirolGÜndeyCÇinarA (2002) A modular simulation package for fed-batch fermentation: penicillin production. Computers & Chemical Engineering26: 1553–1565.
3.
ChaiZZhaoCHHuangB, et al. (2022) A deep probabilistic transfer learning framework for soft sensor modeling with missing data. IEEE Transactions on Neural Networks and Learning Systems33: 7598–7609.
4.
ChenHTHuangB (2023) Fault-tolerant soft sensors for dynamic systems. IEEE Transactions on Control Systems Technology31: 2805–2818.
5.
ChenHTLuoHYHuangB, et al. (2024) Transfer learning-motivated intelligent fault diagnosis designs: A survey, insights, and perspectives. IEEE Transactions on Neural Networks and Learning Systems35: 2969–2983.
6.
ChuFWangJCZhaoX, et al. (2021) Transfer learning for nonlinear batch process operation optimization. Journal of Process Control101: 11–23.
7.
GongBQShiYShaF, et al. (2012) Geodesic flow kernel for unsupervised domain adaptation. In: 2012 IEEE conference on computer vision and pattern recognition, Providence, Rhode Island, USA, 16–21 June 2012, pp. 2066–2073. Piscataway: IEEE.
8.
GrettonABorgwardtKMRaschMJ, et al. (2012) A kernel two-sample test. Journal of Machine Learning Research13: 723–773.
9.
JiangXYGeZQ (2021) Augmented multidimensional convolutional neural network for industrial soft sensing. IEEE Transactions on Instrumentation and Measurement70: 1–10.
10.
JiangXYGeZQ (2022) Improving the performance of just-intime learning-based soft sensor through data augmentation. IEEE Transactions on Industrial Electronics69: 13716–13726.
11.
JiangYCYinSDongJW, et al. (2021) A review on soft sensors for monitoring, control, and optimization of industrial processes. IEEE Sensors Journal21: 12868–12881.
12.
LiDLiuYQHuangDP, et al. (2023) Development of an adversarial transfer learning-based soft sensor in industrial systems. IEEE Transactions on Instrumentation and Measurement72: 3000610.
13.
LiQYangMHLuZY, et al. (2022) A soft-sensing method for product quality monitoring based on particle swarm optimization deep belief networks. Transactions of the Institute of Measurement and Control44: 2900–2910.
14.
LiSLiuCHSuLM, et al. (2020) Discriminative transfer feature and label consistency for cross-domain image classification. IEEE Transactions on Neural Networks and Learning Systems31: 4842–4856.
15.
LiSSongSJHuangG (2017) Prediction reweighting for domain adaptation. IEEE Transactions on Neural Networks and Learning Systems28: 1682–1695.
16.
LiSSongSJHuangG, et al. (2018) Domain invariant and class discriminative feature learning for visual domain adaptation. IEEE Transactions on Image Processing27: 4260–4273.
17.
LiZWangXKrugerU (2021) Efficient cross-validatory algorithm for identifying dynamic nonlinear process models. Control Engineering Practice15: 39–52.
18.
LiuWTHeSDMouJP, et al. (2023) Digital twins-based process monitoring for wastewater treatment processes. Reliability Engineering & System Safety238: 109416.
19.
LiuYYangCZhangMT, et al. (2020) Development of adversarial transfer learning soft sensor for multigrade processes. Industrial & Engineering Chemistry Research59: 16330–16345.
20.
LongMSWangJMDingGG, et al (2013) Transfer feature learning with joint distribution adaptation. In: 2013 IEEE international conference on computer vision, Sydney, NSW, Australia, 1–8 December 2013, pp. 2200–2207. Piscataway: IEEE.
21.
LuWChenYQWangJD, et al. (2021) Cross-domain activity recognition via substructural optimal transport. Neurocomputing454: 65–75.
22.
LuoXLeiQWangHR (2024) Gaussian mixture model sample selection strategy-based active semi-supervised soft sensor for industrial processes. Transactions of the Institute of Measurement and Control46: 962–972.
23.
MahsaBMehrtashTHBrianCL, et al. (2014) Domain adaptation on the statistical manifold. In: 2014 IEEE conference on computer vision and pattern recognition, Columbus, OH, USA, 23–28 June 2014, pp. 2481–2488. Piscataway: IEEE.
24.
Nikzad-LangerodiRZellingerWSaminger-PlatzS, et al. (2021) Domain adaptation for regression under Beer–Lambert’s law. Knowledge-Based Systems210: 106447.
25.
PanHGSuTHuangXD, et al. (2021) LSTM-based soft sensor design for oxygen content of flue gas in coal-fired power plant. Transactions of the Institute of Measurement and Control43: 78–87.
26.
PanSJTsangIWKwokJT, et al. (2011) Domain adaptation via transfer component analysis. IEEE Transactions on Neural Networks22: 199–210.
27.
ShaoWMGeZQSongZH (2021) Semisupervised Bayesian Gaussian mixture models for non-Gaussian soft sensor. IEEE Transactions on Cybernetics51: 3455–3468.
28.
SunBCFengJSSaenkoK (2016) Return of frustratingly easy domain adaptation. In: Proceedings of the thirtieth AAAI conference on artificial intelligence, Phoenix, Arizona, USA, 12–17 February 2016, pp. 2058–2065. Palo Alto: AAAI Press.
29.
SunQQGeZQ (2021) A survey on deep learning for data-driven soft sensors. IEEE Transactions on Industrial Informatics17: 5853–5866.
30.
WangJDChenYQHaoSJ, et al. (2017) Balanced distribution adaptation for transfer learning. In: 2017 IEEE international conference on data mining, New Orleans, LA, USA, 18–21 November 2017, pp. 1129–1134. New York: IEEE.
31.
WangJDFengWJChenYQ (2018) Visual domain adaptation with manifold embedded distribution alignment. In: Proceedings of the 26th ACM international conference on multimedia, Seoul, Republic of Korea, 22–26 October 2018, pp. 402–410. Piscataway: Association for Computing Machinery.
32.
WangJLQiuKPWangRT, et al. (2021) Development of soft sensor based on sequential kernel fuzzy partitioning and just-in-time relevance vector machine for multiphase batch processes. IEEE Transactions on Instrumentation and Measurement70: 1–10.
33.
XuJYCaiZD (2022) Gaussian mixture deep dynamic latent variable model with application to soft sensing for multimode industrial processes. Applied Soft Computing114: 108092.
34.
YaoLShenBBCuiLL, et al. (2023) Semi-supervised deep dynamic probabilistic latent variable model for multimode process soft sensor application. IEEE Transactions on Systems Industrial Informatics19: 6056–6068.
35.
YaoSYKangQZhouMC, et al. (2022) Discriminative manifold distribution alignment for domain adaptation. IEEE Transactions on Systems Man Cybernetics-Systems53: 1183–1197.
36.
YinSLLiYGSunB, et al. (2022) Mixed kernel principal component weighted regression based on just-in-time learning for soft sensor modeling. Measurement Science and Technology33: 015102.
37.
ZhangSNLiuKZhangH, et al. (2024) Dynamic nonlinear soft sensor modelling method using linear slow feature analysis and least squares support vector regression for batch processes. Canadian Journal of Chemical Engineering124: 1–13.
38.
ZhangXRSongCYZhaoJ, et al. (2022) Domain adaptation mixture of Gaussian processes for online soft sensor modeling of multimode processes when sensor degradation occurs. IEEE Transactions on Industrial Informatics18: 4654–4664.
39.
ZhangXRSongCYZhaoJ, et al. (2023) Gaussian mixture continuously adaptive regression for multimode processes soft sensing under time-varying virtual drift. Journal of Process Control124: 1–13.
40.
ZhaoZJYanGWRenMF, et al. (2022) Dynamic transfer partial least squares for domain adaptive regression. Journal of Process Control118: 55–68.
41.
ZhuJLJiaMWZhangY, et al. (2023) Domain adaptation graph convolution network for quality inferring of batch processes. Chemometrics and Intelligent Laboratory Systems243: 105028.
42.
ZhuangFZQiZYDuanKY, et al. (2020) A comprehensive survey on transfer learning. Proceedings of the IEEE109: 43–76.
43.
ZhuangLGaoXB (2024) Transfer adaptation learning: a decade survey. IEEE Transactions on Neural Networks and Learning Systems35: 23–44.
