Electric parameter prediction of rod pumping units based on N-HiTS model

Xiangyu Li, Zhiqing Liu, Chunhua Yuan and Zhupei Liao

PLOS ONE, 2025, vol. 20, issue 7, 1-26

Abstract: The prediction of the operation state of rod pumping unit is one of the important issues in rod pumping engineering. The electrical parameters of the driving motor are gradually becoming the main research direction because of its stability and long-term acquisition. However, there are problems with the existing collection of electrical parameter data, such as a large volume of highly repetitive data within a day, the operation conditions remaining the same for a long time, and fault conditions being rare. This article proposes a new method of combining electrical parameter sequences to predict the operating conditions of rod pumping units. Firstly, an improved mechanism model was established from the rod pumping unit to the driving motor, generating electrical parameter data for typical fault conditions and composite conditions. Secondly, by combining the generated electrical parameter sequences into a time series, the Neural Hierarchical Interpolation for Time Series (N-HiTS) model from the NeuralForecast library is used for prediction. The experimental results show that the proposed method can predict real electrical parameters, generate electrical parameters, and simulate electrical parameters under composite fault conditions. The approach of generating electrical parameters and making predictions can effectively address the issue of insufficient electrical parameter samples, thus enabling a more precise prediction of the operating state of the pumping unit, and adjusting production activities of oil wells.

Date: 2025
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0326973 (text/html)
https://journals.plos.org/plosone/article/file?id= ... 26973&type=printable (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:plo:pone00:0326973

DOI: 10.1371/journal.pone.0326973

Access Statistics for this article

More articles in PLOS ONE from Public Library of Science
Bibliographic data for series maintained by plosone ().