Fault Diagnosis of Pumping Units Based on Extended-Fusion Machine Learning

Lixian Xu; Rui Su; Aftab Ali; Jun Liu; Xinfeng Liu; Dehu Yu

doi:10.1109/iske60036.2023.10480947

Fault Diagnosis of Pumping Units Based on Extended-Fusion Machine Learning

Lixian Xu, Rui Su, Aftab Ali, Jun Liu, Xinfeng Liu, Dehu Yu

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

The failure of a sucker pump in the sucker rod pumping system significantly affects the process of oil production process. To address this issue, this paper proposes a machine learning diagnosis model based on the numerical information fusion of electrical parameters and indicator diagram. The study evaluated the performance of XGBoost and Random Forest models trained on the datasets, with the multi-order constraint conditions set, and the parameters optimized via Adaptive Genetic Algorithm. The experimental results demonstrate that the proposed method achieves accurate and efficient diagnosis of working conditions in pumping oil wells, which significantly improves the working efficiency of pumping oil wells.

Original language	English
Title of host publication	ISKE 2023 - 18th International Conference on Intelligent Systems and Knowledge Engineering
Publisher	IEEE
Pages	302-308
Number of pages	7
ISBN (Electronic)	9798350318401
ISBN (Print)	979-8-3503-1841-8
DOIs	https://doi.org/10.1109/iske60036.2023.10480947
Publication status	Published online - 4 Aug 2024
Event	18th International Conference on Intelligent Systems and Knowledge Engineering - Fuzhou, China Duration: 17 Nov 2023 → 19 Nov 2023 Conference number: 18 http://www.iske2023.com

Publication series

Name	ISKE 2023 - 18th International Conference on Intelligent Systems and Knowledge Engineering

Conference

Conference	18th International Conference on Intelligent Systems and Knowledge Engineering
Abbreviated title	ISKE 2023
Country/Territory	China
City	Fuzhou
Period	17/11/23 → 19/11/23
Internet address	http://www.iske2023.com

Bibliographical note

Publisher Copyright:
© 2023 IEEE.

Keywords

Fault diagnosis
Adaptation models
Oils
pumps
Data models
Real-time systems
Numerical models
indicator diagram
Random forest
XGBoost
Adaptive genetic algorithm
Fault Diagnosis

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/iske60036.2023.10480947

Cite this

Xu, L., Su, R., Ali, A., Liu, J., Liu, X., & Yu, D. (2024). Fault Diagnosis of Pumping Units Based on Extended-Fusion Machine Learning. In ISKE 2023 - 18th International Conference on Intelligent Systems and Knowledge Engineering (pp. 302-308). (ISKE 2023 - 18th International Conference on Intelligent Systems and Knowledge Engineering). IEEE. Advance online publication. https://doi.org/10.1109/iske60036.2023.10480947

@inproceedings{3e84884de3274b50a8fcb02f1959508d,

title = "Fault Diagnosis of Pumping Units Based on Extended-Fusion Machine Learning",

abstract = "The failure of a sucker pump in the sucker rod pumping system significantly affects the process of oil production process. To address this issue, this paper proposes a machine learning diagnosis model based on the numerical information fusion of electrical parameters and indicator diagram. The study evaluated the performance of XGBoost and Random Forest models trained on the datasets, with the multi-order constraint conditions set, and the parameters optimized via Adaptive Genetic Algorithm. The experimental results demonstrate that the proposed method achieves accurate and efficient diagnosis of working conditions in pumping oil wells, which significantly improves the working efficiency of pumping oil wells.",

keywords = "Fault diagnosis, Adaptation models, Oils, pumps, Data models, Real-time systems, Numerical models, indicator diagram, Random forest, XGBoost, Adaptive genetic algorithm, Fault Diagnosis",

author = "Lixian Xu and Rui Su and Aftab Ali and Jun Liu and Xinfeng Liu and Dehu Yu",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 18th International Conference on Intelligent Systems and Knowledge Engineering, ISKE 2023 ; Conference date: 17-11-2023 Through 19-11-2023",

year = "2024",

month = aug,

day = "4",

doi = "10.1109/iske60036.2023.10480947",

language = "English",

isbn = "979-8-3503-1841-8",

series = "ISKE 2023 - 18th International Conference on Intelligent Systems and Knowledge Engineering",

publisher = "IEEE",

pages = "302--308",

booktitle = "ISKE 2023 - 18th International Conference on Intelligent Systems and Knowledge Engineering",

address = "United States",

url = "http://www.iske2023.com",

}

Xu, L, Su, R, Ali, A , Liu, J, Liu, X & Yu, D 2024, Fault Diagnosis of Pumping Units Based on Extended-Fusion Machine Learning. in ISKE 2023 - 18th International Conference on Intelligent Systems and Knowledge Engineering. ISKE 2023 - 18th International Conference on Intelligent Systems and Knowledge Engineering, IEEE, pp. 302-308, 18th International Conference on Intelligent Systems and Knowledge Engineering, Fuzhou, China, 17/11/23. https://doi.org/10.1109/iske60036.2023.10480947

Fault Diagnosis of Pumping Units Based on Extended-Fusion Machine Learning. / Xu, Lixian; Su, Rui; Ali, Aftab et al.
ISKE 2023 - 18th International Conference on Intelligent Systems and Knowledge Engineering. IEEE, 2024. p. 302-308 (ISKE 2023 - 18th International Conference on Intelligent Systems and Knowledge Engineering).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Fault Diagnosis of Pumping Units Based on Extended-Fusion Machine Learning

AU - Xu, Lixian

AU - Su, Rui

AU - Ali, Aftab

AU - Liu, Jun

AU - Liu, Xinfeng

AU - Yu, Dehu

N1 - Conference code: 18

PY - 2024/8/4

Y1 - 2024/8/4

N2 - The failure of a sucker pump in the sucker rod pumping system significantly affects the process of oil production process. To address this issue, this paper proposes a machine learning diagnosis model based on the numerical information fusion of electrical parameters and indicator diagram. The study evaluated the performance of XGBoost and Random Forest models trained on the datasets, with the multi-order constraint conditions set, and the parameters optimized via Adaptive Genetic Algorithm. The experimental results demonstrate that the proposed method achieves accurate and efficient diagnosis of working conditions in pumping oil wells, which significantly improves the working efficiency of pumping oil wells.

AB - The failure of a sucker pump in the sucker rod pumping system significantly affects the process of oil production process. To address this issue, this paper proposes a machine learning diagnosis model based on the numerical information fusion of electrical parameters and indicator diagram. The study evaluated the performance of XGBoost and Random Forest models trained on the datasets, with the multi-order constraint conditions set, and the parameters optimized via Adaptive Genetic Algorithm. The experimental results demonstrate that the proposed method achieves accurate and efficient diagnosis of working conditions in pumping oil wells, which significantly improves the working efficiency of pumping oil wells.

KW - Fault diagnosis

KW - Adaptation models

KW - Oils

KW - pumps

KW - Data models

KW - Real-time systems

KW - Numerical models

KW - indicator diagram

KW - Random forest

KW - XGBoost

KW - Adaptive genetic algorithm

KW - Fault Diagnosis

UR - http://www.scopus.com/inward/record.url?scp=85191017930&partnerID=8YFLogxK

U2 - 10.1109/iske60036.2023.10480947

DO - 10.1109/iske60036.2023.10480947

M3 - Conference contribution

SN - 979-8-3503-1841-8

T3 - ISKE 2023 - 18th International Conference on Intelligent Systems and Knowledge Engineering

SP - 302

EP - 308

BT - ISKE 2023 - 18th International Conference on Intelligent Systems and Knowledge Engineering

PB - IEEE

T2 - 18th International Conference on Intelligent Systems and Knowledge Engineering

Y2 - 17 November 2023 through 19 November 2023

ER -

Xu L, Su R, Ali A , Liu J, Liu X, Yu D. Fault Diagnosis of Pumping Units Based on Extended-Fusion Machine Learning. In ISKE 2023 - 18th International Conference on Intelligent Systems and Knowledge Engineering. IEEE. 2024. p. 302-308. (ISKE 2023 - 18th International Conference on Intelligent Systems and Knowledge Engineering). Epub 2024 Aug 4. doi: 10.1109/iske60036.2023.10480947

Fault Diagnosis of Pumping Units Based on Extended-Fusion Machine Learning

Abstract

Publication series

Conference

Bibliographical note

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this