智能化矿井提升系统健康监测关键技术

Key technologies for health monitoring of intelligent mine hoisting systems

  • 摘要: 目前矿井提升系统健康监测存在以下问题:系统跨度大、部件分布离散,难以实现对关键部件的全域覆盖、连续在线与高性价比监测部署;关键部件故障早期劣化往往表现为弱特征,在多源扰动与强噪声背景下,微弱故障信息易被掩盖,导致特征提取难度显著增加;服役工况恶劣、可靠性要求高。介绍了矿井提升系统的结构组成及功能,重点分析了滚筒、主轴、钢丝绳、轴承等关键部件在长期服役过程中对健康状态监测的核心需求。在此基础上,详细阐述了当前矿井提升系统健康监测中的2类核心感知技术:一类是基于振动、声音、视觉、温度等多元监测信号的定点感知技术,阐明了各类信号的采集方式、感知原理及适用场景;另一类是基于移动机器人巡检、提升容器集成式巡检的移动感知技术,分析了各技术特点、运行流程及应用局限。同时,分析了健康状态评估方法在矿井提升系统各关键部件状态监测中的应用,探讨了基于信号处理、机器学习、深度学习的智能评估方法原理、效果及特点。总结了传统矿井提升系统监测平台与数字孪生平台的发展现状及技术特点。基于当前矿井提升系统健康监测技术存在的问题与挑战,指出矿井提升系统应从复杂工况感知优化、移动感知智能化、评估模型高效化、监测系统融合化等方面发展。

     

    Abstract: At present, health monitoring of mine hoisting systems faces challenges such as large system span and dispersed component distribution, which makes it difficult to achieve full coverage, continuous online monitoring, and cost-effective deployment for key components. Early degradation of key components often manifests as weak features. Under multi-source disturbances and strong noise backgrounds, weak fault information is easily masked, significantly increasing the difficulty of feature extraction. In addition, harsh service conditions and high reliability requirements further complicate monitoring tasks. This study introduces the structural composition and functions of mine hoisting systems, with a focus on analyzing the core requirements for health state monitoring of key components such as the drum, main shaft, steel wire rope, and bearings during long-term service. On this basis, two types of core sensing technologies used in current mine hoisting system health monitoring are described in detail. One type is fixed-point sensing technology based on multiple monitoring signals such as vibration, sound, vision, and temperature, and the acquisition methods, sensing principles, and applicable scenarios of each type of signal are explained. The other type is mobile sensing technology based on inspection using mobile robots and integrated inspection of hoisting conveyances, and the characteristics, operating processes, and application limitations of these technologies are analyzed. In addition, the application of health state assessment methods in the state monitoring of key components of mine hoisting systems is analyzed, and the principles, effectiveness, and characteristics of intelligent assessment methods based on signal processing, machine learning, and deep learning are discussed. The development status and technical characteristics of traditional mine hoisting system monitoring platforms and digital twin platforms are summarized. Based on the existing problems and challenges in current health monitoring technologies for mine hoisting systems, future development directions are proposed, including optimization of sensing under complex operating conditions, intelligent mobile sensing, efficient evaluation models, and integrated monitoring systems.

     

/

返回文章
返回