Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Regression generation adversarial network based on dual data evaluation strategy for industrial application

Zesen Wang, Yonggang Li, Lijuan Lan

TL;DR

This work tackles data scarcity in industrial soft sensing by introducing RGAN-DDE, a regression-aware GAN with a shallow regressor-discriminator sharing mechanism and a dual data evaluation strategy. By embedding label-feature mappings into both generator and discriminator and actively curating training and generated samples, the method generates more diverse, distribution-consistent data and improves downstream soft-sensing models. Across four industrial datasets, RGAN-DDE outperforms state-of-the-art generative approaches in accuracy and convergence, while maintaining competitive efficiency. The approach offers practical benefits for reducing labeling needs and enhancing generalization in real-world industrial settings.

Abstract

Soft sensing infers hard-to-measure data through a large number of easily obtainable variables. However, in complex industrial scenarios, the issue of insufficient data volume persists, which diminishes the reliability of soft sensing. Generative Adversarial Networks (GAN) are one of the effective solutions for addressing insufficient samples. Nevertheless, traditional GAN fail to account for the mapping relationship between labels and features, which limits further performance improvement. Although some studies have proposed solutions, none have considered both performance and efficiency simultaneously. To address these problems, this paper proposes the multi-task learning-based regression GAN framework that integrates regression information into both the discriminator and generator, and implements a shallow sharing mechanism between the discriminator and regressor. This approach significantly enhances the quality of generated samples while improving the algorithm's operational efficiency. Moreover, considering the importance of training samples and generated samples, a dual data evaluation strategy is designed to make GAN generate more diverse samples, thereby increasing the generalization of subsequent modeling. The superiority of method is validated through four classic industrial soft sensing cases: wastewater treatment plants, surface water, $CO_2$ absorption towers, and industrial gas turbines.

Regression generation adversarial network based on dual data evaluation strategy for industrial application

TL;DR

This work tackles data scarcity in industrial soft sensing by introducing RGAN-DDE, a regression-aware GAN with a shallow regressor-discriminator sharing mechanism and a dual data evaluation strategy. By embedding label-feature mappings into both generator and discriminator and actively curating training and generated samples, the method generates more diverse, distribution-consistent data and improves downstream soft-sensing models. Across four industrial datasets, RGAN-DDE outperforms state-of-the-art generative approaches in accuracy and convergence, while maintaining competitive efficiency. The approach offers practical benefits for reducing labeling needs and enhancing generalization in real-world industrial settings.

Abstract

Soft sensing infers hard-to-measure data through a large number of easily obtainable variables. However, in complex industrial scenarios, the issue of insufficient data volume persists, which diminishes the reliability of soft sensing. Generative Adversarial Networks (GAN) are one of the effective solutions for addressing insufficient samples. Nevertheless, traditional GAN fail to account for the mapping relationship between labels and features, which limits further performance improvement. Although some studies have proposed solutions, none have considered both performance and efficiency simultaneously. To address these problems, this paper proposes the multi-task learning-based regression GAN framework that integrates regression information into both the discriminator and generator, and implements a shallow sharing mechanism between the discriminator and regressor. This approach significantly enhances the quality of generated samples while improving the algorithm's operational efficiency. Moreover, considering the importance of training samples and generated samples, a dual data evaluation strategy is designed to make GAN generate more diverse samples, thereby increasing the generalization of subsequent modeling. The superiority of method is validated through four classic industrial soft sensing cases: wastewater treatment plants, surface water, absorption towers, and industrial gas turbines.
Paper Structure (14 sections, 9 equations, 7 figures, 4 tables, 2 algorithms)

This paper contains 14 sections, 9 equations, 7 figures, 4 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Related work
  3. GAN, WGAN-GP, and MR-GAN
  4. GAN for Soft Sensor
  5. Propoesd method
  6. RGAN
  7. Dual Data Evaluation Strategy
  8. Case study
  9. Dataset
  10. Experiment Settings
  11. Main Results
  12. Ablation Experiment
  13. Model Analysis
  14. Conclusion

Figures (7)

  • Figure 1: Structure of GAN.
  • Figure 2: Flowchart of GAN for soft sensor.
  • Figure 3: Structure of RGAN-DDE
  • Figure 4: Visualization analysis. Boxplot comparison of real and generated feature distributions for different GAN models.
  • Figure 5: Visualization analysis. Scatter plot comparison of real and generated samples for four cases under different GAN-based augmentation methods.
  • ...and 2 more figures