基于YOLOv11的风电机组主轴螺栓状态智能监测系统  

Condition Monitoring System for Main Shaft Bolts in Wind Turbines Based on YOLOv11

作　　者：陈怀 陈亚楠 李籽圆 张家友 覃龙 舒晖 CHEN Huai;CHEN Yanan;LI Ziyuan;ZHANG Jiayou;QIN Long;SHU Hui(CRRC Zhuzhou Institute Co.,Ltd.,Zhuzhou,Hunan 412001,China)

机构地区：[1]中车株洲电力机车研究所有限公司,湖南株洲412001

出　　处：《控制与信息技术》2025年第1期27-32,共6页CONTROL AND INFORMATION TECHNOLOGY

摘　　要：主轴螺栓的异常状态监测在保证风电机组的稳定和高效运行中起着至关重要的作用。传统的人工巡检方法存在效率低、实时性差的问题,难以满足现代风电机组高效运维的要求;基于传感器的监测方法虽然能够实现自动化检测,但存在功能单一、运维成本高等问题,不易于大规模应用。为实现风电机组主轴螺栓状态的智能化监测,文章提出一种基于YOLOv11的主轴螺栓状态监测方法。首先,使用YOLOv11算法检测主轴螺栓的多种异常情况,包括螺母脱落、螺栓松动、螺栓掉落等;然后,将识别到的异常状态信息集成并展示在风电机组的SCADA系统界面上,提醒操作人员及时采取相应的维护措施,以确保机组的安全运行。实验结果证明,本文方法检测精度达到了96.2%,且推理速度能够达到400帧/秒,能够满足风机高效运维的要求。Abnormal condition monitoring of main shaft bolts plays a crucial role in ensuring the stable and efficient operation of wind turbines.The traditional inspection method of manual patrols has proven inefficient and lacking in real-time performance,which hinders its ability to meet the requirements of efficient operation and maintenance for modern wind turbines.Although sensorbased monitoring methods support automated detection,they are restricted from large-scale applications due to their simple functionality and high operational and maintenance costs.This paper proposes a YOLOv11-based system to facilitate intelligent condition monitoring for main shaft bolts.First,the YOLOv11 algorithm is implemented to detect a variety of abnormal conditions of main shaft bolts,including nut detachment,bolt loosening,and bolt falling-off.Next,the information regarding the recognized abnormal conditions is integrated and displayed on the SCADA system interface of the wind turbines,which reminds operators to take corresponding maintenance measures in time,thereby ensuring safe operation.The experimental demonstrated the proposed method achieved a detection accuracy of up to 96.2%and a detection frame rate of 400 FPS,aligning with the requirements of efficient operation and maintenance for wind turbines.

关 键 词：风电机组 机器视觉 主轴螺栓 状态监测 YOLOv11 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术] TM614[自动化与计算机技术—计算机科学与技术]