A Diffusion Model-Empowered CNN-Transformer for Few-Shot Fault Diagnosis in Natural Gas Wells

Wang, Chuanping; Li, Yudong; Wang, Jiajia; Wang, Yuzhe; Liu, Yufeng; Han, Ling; Yang, Fan; Gao, Xiaoyong

doi:10.3390/pr13082608

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

A Diffusion Model-Empowered CNN-Transformer for Few-Shot Fault Diagnosis in Natural Gas Wells

by

Chuanping Wang

¹,

Yudong Li

¹,

Jiajia Wang

¹,

Yuzhe Wang

¹,

Yufeng Liu

¹,

Ling Han

¹,

Fan Yang

¹ and

Xiaoyong Gao

^2,*

¹

No.1 Gas Production Plant, PetroChina Xinjiang Oilfield Company, Karamay 834000, China

²

I3Lab, Department of Automation, China University of Petroleum, Beijing 102249, China

^*

Author to whom correspondence should be addressed.

Processes 2025, 13(8), 2608; https://doi.org/10.3390/pr13082608

Submission received: 12 July 2025 / Revised: 6 August 2025 / Accepted: 14 August 2025 / Published: 18 August 2025

(This article belongs to the Special Issue Progress in Design and Optimization of Fault Diagnosis Modelling)

Download Versions Notes

Abstract

Natural gas wells operate under complex conditions with frequent environmental disturbances. Fault types vary significantly and often present weak signals, affecting both safety and efficiency. This paper proposes an intelligent fault-diagnosis method based on a CNN-Transformer model using real-time wellsite data. A time series diffusion model is applied to enhance small-sample data by generating synthetic fault samples, and the CNN-Transformer model extracts both local and global features from time series inputs to improve fault recognition in complex scenarios. Validation on a real-world dataset demonstrates that the proposed method achieves a macro F1-Score of 99.52% in multi-class fault diagnosis, significantly outperforming baseline models (1D-CNN: 95.83%, LSTM: 93.54%, GRU: 94.98%). Quantitative analysis confirms the diffusion model’s superiority in data augmentation, with lower Earth Mover’s Distance (0.087), KL Divergence (0.245), and Mean Squared Error (0.298) compared to GAN and VAE variants. Ablation studies show that removing diffusion-based augmentation leads to a 14.96% drop in F1-Score, highlighting its critical role in mitigating class imbalance. Results validate the diffusion model’s effectiveness for data augmentation and the CNN-Transformer’s superior ability to capture complex time series patterns, providing theoretical support and practical tools for intelligent monitoring and maintenance in natural gas well systems.

Keywords: natural gas wells; fault diagnosis; diffusion modeling; CNN; transformer; time series

Share and Cite

MDPI and ACS Style

Wang, C.; Li, Y.; Wang, J.; Wang, Y.; Liu, Y.; Han, L.; Yang, F.; Gao, X. A Diffusion Model-Empowered CNN-Transformer for Few-Shot Fault Diagnosis in Natural Gas Wells. Processes 2025, 13, 2608. https://doi.org/10.3390/pr13082608

AMA Style

Wang C, Li Y, Wang J, Wang Y, Liu Y, Han L, Yang F, Gao X. A Diffusion Model-Empowered CNN-Transformer for Few-Shot Fault Diagnosis in Natural Gas Wells. Processes. 2025; 13(8):2608. https://doi.org/10.3390/pr13082608

Chicago/Turabian Style

Wang, Chuanping, Yudong Li, Jiajia Wang, Yuzhe Wang, Yufeng Liu, Ling Han, Fan Yang, and Xiaoyong Gao. 2025. "A Diffusion Model-Empowered CNN-Transformer for Few-Shot Fault Diagnosis in Natural Gas Wells" Processes 13, no. 8: 2608. https://doi.org/10.3390/pr13082608

APA Style

Wang, C., Li, Y., Wang, J., Wang, Y., Liu, Y., Han, L., Yang, F., & Gao, X. (2025). A Diffusion Model-Empowered CNN-Transformer for Few-Shot Fault Diagnosis in Natural Gas Wells. Processes, 13(8), 2608. https://doi.org/10.3390/pr13082608

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.

Article Menu

A Diffusion Model-Empowered CNN-Transformer for Few-Shot Fault Diagnosis in Natural Gas Wells

Abstract

Share and Cite

Article Metrics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI