Semi-supervised soft sensor method for fermentation processes based on physical monotonicity and variational autoencoders

• Published in: Engineering applications of artificial intelligence 2024-11, Vol.137, p.109065, Article 109065
• Main Authors: Cheng, Xinyue, Yu, Zhenhua, Wang, Guan, Jiang, Qingchao, Cao, Zhixing
• Format: Article
• Language: English
• Subjects: Fermentation process; Semi-supervised learning; Soft sensor; Variational autoencoder
• ISSN: 0952-1976
• DOI: 10.1016/j.engappai.2024.109065

Data-driven models have shown broad application prospects in soft sensor modeling. However, numerous challenges persist. On the one hand, data-driven soft sensor methods have high requirements on data quality. On the other hand, models relying on limited experimental data often lack physical interpretability. To tackle these challenges, a semi-supervised soft sensor method (PMVAER) for fermentation processes based on physical monotonicity and variational autoencoders (VAEs) is introduced. First, physical monotonicity constraint is incorporated into the loss function of VAEs for regression to ensure that the model's predictions adhere to physical feasibility. Next, considering the disparate sampling frequencies for process and quality variables, this approach is extended to learn from unlabeled data, creating a semi-supervised soft sensor model. The proposed model is validated on simulation and real cases of penicillin fermentation. Comparisons with five other methods verify that the proposed method exhibits exceptional predictive accuracy along with enhanced generalization ability. [Display omitted] •A novel PMVAER model was developed for soft sensing of penicillin fermentation.•The PMVAER model integrated experimental data and physical monotonicity.•The model disentangled the dimension related to labels from the latent space.•The PMVAER model showed superior prediction performance (higher R2 and lower RMSE).•The PMVAER model showed better interpretability and stronger generalization ability.