Special Oil & Gas Reservoirs ›› 2025, Vol. 32 ›› Issue (3): 133-141.DOI: 10.3969/j.issn.1006-6535.2025.03.016

• Drilling & Production Engineering • Previous Articles     Next Articles

Fatigue life prediction of sheave wheels for ultra-deep well drilling rigs

YI Xianzhong1,2, QIN Saibo1, LIU Hangming3, YUE Rongchang1, XU Zhixin1, PI Yunsong4, CAI Xingxing4, JI Yusong4   

  1. 1. School of Mechanical Engineering, Yangtze University, Jingzhou, Hubei 434023, China;
    2. Hubei Provincial Enterprise and University Joint Innovation Center for Intelligent Oil and Gas Drilling and Production Equipment, Jingzhou, Hubei 434000, China;
    3. The Seventh Geological Brigade of Hubei Geological Bureau, Yichang, Hubei 443000, China;
    4. Hubei Jianghan Petroleum Instrument & Meter Co., Ltd., Wuhan, Hubei 430205, China
  • Received:2024-09-22 Revised:2025-02-05 Online:2025-06-25 Published:2025-07-08

Abstract: To address the issue of unclear understanding of the fatigue life of critical components in the drilling rig hoisting system, specifically the deadline anchor, the fatigue life analysis was conducted using the finite element method, taking the JZG97 type deadline anchor used in the ZJ150 drilling rig as an example. Through orthogonal experiments, the structural parameters of the weak link (sheave wheel) of the deadline anchor were optimized. Furthermore, a BP neural network fatigue life prediction model was established to predict the fatigue life of the vulnerable parts. The results show that the optimal parameters are a rib width of 90 mm, a sensor arm fillet radius of 270 mm, and a rib thickness of 230 mm. This increases the sheave wheel′s fatigue safety factor to 1.78, a 19.46% improvement over the initial design. The BP neural network model has a fitting accuracy of 97.84% and a maximum error of less than 4.870% between the predicted value and the simulated value. It proves to be accurate, fast-converging, and generalizable, enabling reliable fatigue life prediction of the deadline anchor without constructing complex functional regression relationships. This research offers guidance for structural optimization and life prediction of deadline anchors in ultra-deep well drilling equipment.

Key words: deadline anchor, sheave wheel, fatigue life, orthogonal experiment, BP neural network

CLC Number: