Fault Diagnosis of Pumping Units Based on Multi-scale Feature Extraction and ResNet-Transformer

HAN Dongying; ZHU Zhizhou; GE Zixuan; SHI Peiming

doi:10.3969/j.issn.1000-1158.2026.01.05

PDF(2652 KB)

Acta Metrologica Sinica ›› 2026, Vol. 47 ›› Issue (1) : 35-41. DOI: 10.3969/j.issn.1000-1158.2026.01.05

Fault Diagnosis of Pumping Units Based on Multi-scale Feature Extraction and ResNet-Transformer

Author information +

History +

Abstract

A multi-scale feature extraction and ResNet-Transformer algorithm （MSFRT） is proposed for fault diagnosis of pumping units. Firstly， the local feature extraction capability of the deep residual network ResNet-34 is utilized to capture the spatial details of dynamometer cards， and the context modeling capability of the Transformer encoder is utilized to obtain global features， constructing an end-to-end fault diagnosis framework for pumping units； Secondly， a multi-scale feature extraction module is introduced， which extracts feature information of different scales in parallel through 1×1， 3×3， and 5×5 convolutional kernels， enhancing the ability to perceive details in dynamometer cards； Finally， a feature fusion attention mechanism is designed to adaptively integrate multi-scale features and global semantic information. Experiments are conducted on a dynamometer cards dataset containing 7 typical operating conditions， and the results show that the algorithm achieved an accuracy of 94% in fault diagnosis tasks， verifying the effectiveness of the proposed algorithm.

Key words

mechanics metrology / fault diagnosis / pumping unit / dynamometer cards / multi-scale feature extraction / ResNet-Transformer model

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

HAN Dongying , ZHU Zhizhou , GE Zixuan , et al. Fault Diagnosis of Pumping Units Based on Multi-scale Feature Extraction and ResNet-Transformer[J]. Acta Metrologica Sinica. 2026, 47(1): 35-41 https://doi.org/10.3969/j.issn.1000-1158.2026.01.05

References

List( Publishing order | Descend order by publishing year | Descend order by cited within ) Chart analysis

[1]	SONG C H， LIU S， HAN G J， et al. Edge-intelligence-based condition monitoring of beam pumping units under heavy noise in industrial internet of things for industry 4.0［J］. IEEE Internet of Things Journal， 2023， 10（4）： 3037–3046. Cited in this article [1]

[2]	WEI W， XIN X， HANG X W， et al. Fault diagnosis of pumping system based on multimodal attention learning （CBMA learning）［J］. Journal of Process Control， 2023， 128：103006. Cited in this article [1]

[3]	LÜ X X， WANG H X， XIN Z， et al. Adaptive fault diagnosis of sucker rod pump systems based on optimal perceptron and simulation data［J］. Petroleum Science， 2022， 19（2）： 743–760. Cited in this article [1]

[4]	WANG X， HE Y F， LI F J， et al. A working condition diagnosis model of sucker rod pumping wells based on deep learning［J］. SPE Production & Operations， 2021， 36（2）： 317–326.

[5]	WANG X， LIU Y， CHENG X， et al. ParaDC： parallel-learning-based dynamometer cards augmentation with diffusion models in sucker rod pump systems［J］. Neurocomputing， 2025， 617： 128973. Cited in this article [1]

[6]	YIN C Z， ZHANG K， ZHANG L M， et al. Imbalanced working states recognition of sucker rod well dynamometer cards based on data generation and diversity augmentation［J］. SPE Journal， 2023， 28（4）： 1925–1944. Cited in this article [1]

[7]	ZHANG A， GAO X W. Supervised dictionary-based transfer subspace learning and applications for fault diagnosis of sucker rod pumping systems［J］. Neurocomputing， 2019， 338： 293–306. Cited in this article [1]

[8]	LÜ X X， FENG L， WANG H X， et al. Quantitative diagnosis method of the sucker rod pump system based on the fault mechanism and inversion algorithm［J］. Journal of Process Control， 2021， 104： 40–53. Cited in this article [1]

[9]	LING H， LI X M. Accurate extraction of valve opening and closing points based on the physical meaning of surface dynamometer card［J］. Petroleum Exploration and Development， 2011， 38（1）： 109–115. Cited in this article [1]

[10]	刘芑辰，冯子明，蒋国斌，等. 基于卷积神经网络模型的抽油机系统故障诊断［J］. 石油钻采工艺， 2021， 43（6）： 777-781. LIU C Y， FENG Z M， JIANG G B， et al. Fault diagnosis of pumping unit system based on convolutional neural network model［J］. Oil Drilling and Production Technology， 2021， 43（6）： 777-781. Cited in this article [1]

[11]	HU H T， LI M， DANG C. Research on the fault identification method of oil pumping unit based on residual network［C］//2022 7th International Conference on Intelligent Computing and Signal Processing （ICSP）. Xi’an， China. 2022： 940–943. Cited in this article [1]

[12]	MA R， TIAN H， CHENG X， et al. Oil-net： a learning-based framework for working conditions diagnosis of oil well through dynamometer cards identification［J］. IEEE Sensors Journal， 2023， 23（13）： 14406–14417. Cited in this article [1]

[13]	ZHANG L， WU J L， ZHANG K， et al. Diagnosis of pumping machine working conditions based on transfer learning and vit model［J］. Geoenergy Science and Engineering， 2023， 226： 211729. Cited in this article [1]

[14]	HAN C J， ZHOU X L， FAN C M， et al. Research on fault analysis model of lightweight pumping unit based on classical convolutional neural network［J］. Journal of Failure Analysis and Prevention， 2023， 23（6）： 2402–2415. Cited in this article [1]

[15]	刘月峰，刘好峰，王越，等. 基于改进Transformer模型的运动想象脑电分类方法研究［J］. 计量学报， 2023， 44（7）： 1147–1153. LIU Y F， LIU H F， WANG Y， et al. Research on EEG classification of motion imagination based on improved Transformer model［J］. Acta Metrologica Sinica， 2023， 44（7）： 1147-1153. Cited in this article [1]

[16]	WANG D C， YANG R H， ZHANG Z X， et al. P-swin： parallel swin transformer multi-scale semantic segmentation network for land cover classification［J］. Computers & Geosciences， 2023， 175： 105340. Cited in this article [1]

[17]	WEI C Y， HAN H， WU Z C， et al. Transformer-based multiscale reconstruction network for defect detection of infrared images［J］. IEEE Transactions on Instrumentation and Measurement， 2024， 73： 1–14. Cited in this article [1]

[18]	刘晨宇，李志农，熊鹏伟，等. 融合路径聚合网络的Swin Transformer的故障诊断方法研究［J］. 振动与冲击， 2024， 43（18）： 258-266. LIU C Y， LI Z N， XIONG P W， et al. Research on Fault Diagnosis Method of Swin Transformer with Fusion Path Aggregation Network［J］. Journal of Vibration and Shock， 2024， 43（18）： 258-266.

[19]	金宇锋，陶重犇. 基于Transformer的融合信息增强3D目标检测算法［J］. 仪器仪表学报， 2023， 44（12）： 297-306. JIN Y F， TAO C B. Fusion Information Enhanced 3D Object Detection Algorithm Based on Transformer［J］. Chinese Journal of Scientific Instrument， 2023， 44（12）： 297-306. Cited in this article [1]