Enhancing the safety of hydroelectric power generation systems: an intelligent identification of axis orbits based on a nonlinear dynamics method

Fei Chen, Zhigao Zhao, Xiaoxi Hu, Dong Liu, Zhe Kang, Zhe Ma, Pengfei Xiao, Xiuxing Yin and Jiandong Yang

Energy, 2025, vol. 324, issue C

Abstract: Hydropower has the highest percentage among renewable energies, and guaranteeing the safety of hydroelectric power generation system is of great significance in promoting the stable operation of the power grid. The axis orbit is an important index in the monitoring of hydraulic turbines. Different shapes characterize the various operating statuses of the hydraulic turbine, and accurately identifying these shapes has been a crucial issue in the intelligent operation and maintenance of hydropower plants. However, existing image-based axis orbit identification methods suffer from defects such as poor feature interpretability and weak noise immunity, making their strategy of extracting feature information solely from images unsuitable for complex operating environments. Therefore, this paper returns to the origin of the axis orbit and proposes an intelligent identification method for axis orbits based on swing signals of hydraulic turbine. Firstly, operational data of the axis orbit is collected using an eddy current sensor installed on the shaft system of the hydraulic turbine, providing a set of orthogonal swing signals. Secondly, a new nonlinear dynamic method named refined composite multivariate multiscale dispersion sample entropy (RCMvMDSE) is proposed based on multidimensional embedding theory. Finally, random forest (RF) and RCMvMDSE are utilized to achieve intelligent identification of the axis orbit. In this paper, the proposed method is applied to three scenarios: simulation, experimentation, and prototype power station. Comparative experiments are then conducted using image recognition techniques and popular nonlinear dynamics methods. The results show that the proposed method achieves excellent identification across all scenarios, with the accuracy rate, precision rate, recall rate, and F1-score of at least 90 %, which is higher than other methods, thereby verifying its advantages. It effectively reduces the likelihood of accidental shutdowns in hydroelectric power generation systems and enhances the stability of power station.

Keywords: Hydraulic turbine; axis orbit; Intelligent identification; Nonlinear dynamics; feature extraction (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544225015063
Full text for ScienceDirect subscribers only

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:eee:energy:v:324:y:2025:i:c:s0360544225015063

DOI: 10.1016/j.energy.2025.135864

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().