TH.2026.02.017]
点击复制

基于振动传感器的煤矿井下带式输送机齿轮箱远程故障诊断系统研究()

《工业仪表与自动化装置》[ISSN:1000-0682/CN:61-1121/TH]

Title:: Research on remote fault diagnosis system for gearbox of underground belt conveyor in coal mines based on vibration sensors

作者:: 郑勇1; 冯彦军2; 3*; 1. 陕西陕煤曹家滩矿业有限公司，陕西榆林，719100； 2.中煤科工开采研究院有限公司，北京 100013； 3.天地科技股份有限公司，北京 100013

Author(s):: ZHENG Yong1; FENG Yanjun2; 3*; 1.Shaanxi Shaanxi Coal Caojiatan Mining Co., Ltd., Shaanxi Yulin 719100, China; 2. China Coal Science and Industry Mining Research Institute Co., Ltd., Beijing 100013,China; 3. Tiandi Technology Co., Ltd., Beijing 100013, China

Keywords:: belt conveyor; gearbox; fault diagnosis; vibration signal; particle swarm optimization

摘要:: 煤矿井下带式输送机齿轮箱故障诊断对保障安全连续生产具有重要意义，然而现有方法多针对输送机轴承而较少涉及齿轮箱，导致故障识别准确率较低且实时性差。基于此，该文提出了基于振动传感的煤矿井下带式输送机齿轮箱远程故障诊断系统，可实现对齿轮断齿、轴承损伤及轴系失衡故障的分类识别。系统通过振动传感器采集输送机齿轮箱数据，并经4G/5G网络传输至地表进行算法处理。算法采用分层阈值降噪技术消除噪声干扰，基于谱峭度定位共振频带并提取包络谱特征，最终通过粒子群优化的决策树支持向量机模型完成故障分类。进一步的现场试验表明系统识别故障的平均准确率为95.9%，满足煤矿的实际使用需求。相关成果可为煤矿带式输送机的安全运行提供必要的技术支撑。

Abstract:: Fault diagnosis of gearboxes in underground belt conveyors is crucial for ensuring safe and continuous coal mine production. However, current methods primarily target conveyor bearings rather than gearboxes, resulting in low fault recognition accuracy and poor real-time performance. To address this, this study proposes a vibration sensor-based remote fault diagnosis system for underground belt conveyor gearboxes, enabling classification of gear tooth fracture, bearing damage, and shaft imbalance faults. The system collects gearbox data via vibration sensors and transmits it to the surface through 4G/5G networks for algorithmic processing. The algorithm employs?hierarchical threshold denoising?to eliminate noise interference,?locates resonance bands via spectral kurtosis?to extract envelope spectrum features, and ultimately classifies faults using a?decision tree-based support vector machine optimized by particle swarm optimization (PSO). Field tests demonstrate an average fault identification accuracy of 95.9%, meeting practical coal mine requirements. These findings provide essential technical support for the safe operation of belt conveyors in coal mines.

[1]刘东. 基于机器视觉的矿井带式输送机节能调速控制系统设计 [J]. 西部探矿工程, 2025, 37 (10): 182-185.

[2]王锐,冯彦军,任建超,等. 基于机器视觉的煤矿井下带式输送机堆煤故障监测系统研发 [J]. 煤矿机械, 2024, 45 (09): 157-159.

?3]徐志军. 煤矿带式输送机变速箱齿轮故障诊断技术研究 [J]. 矿业装备, 2024, (09): 167-169.

[4]祁瑞敏,张国栋,代皓轩. 基于改进模糊信息融合的煤矿带式输送机健康诊断 [J]. 电子测试, 2024, (01): 16-20.

[5]麻荣福,宋泓炎,庄巍,等. 基于ICEEMDAN-GOA-LSSVM的带式输送机托辊轴承故障诊断方法 [J]. 矿山机械, 2025, 53 (04): 18-24.

[6]张硕. 基于AI技术的矿用带式输送机驱动装置故障诊断预警研究 [J]. 煤矿机电, 2023, 44 (05): 59-63.

[7]权国辉,邰金华,张庆莉,等. 用于矿山皮带输送机滚动轴承故障识别的Xception-CNN模型 [J]. 金属矿山, 2025, (10): 149-158.

[8]江帆,程舒曼,朱真才,等. 基于CLSRIME-XGBOOST的带式输送机托辊故障诊断方法 [J]. 振动、测试与诊断, 2025, 45 (04): 666-673.

?9]秦翥. 分布式光纤传感带式输送机故障诊断系统研究 [J]. 煤矿机械, 2025, 46 (08): 190-192.

[10]陈湘源,秦伟,刘晏驰,等. 融合卷积神经网络与线性回归的带式输送机托辊故障音频识别方法 [J]. 煤炭科学技术, 2025, 53 (S1): 389-398.

[11]王艳. 基于声音阵列和图像处理的皮带机跑偏检测系统 [J]. 测控技术, 2022, 41 (06): 19-25.?/div>

[12]王浩宇,王希鹏. 基于机器视觉的带式输送机胶带跑偏检测方法研究 [J]. 中国煤炭, 2024, 50 (S2): 158-163.

[13]申龙,单浩然,裴文良,等. 基于卷积自编码器的煤矿带式输送机异常声音检测方法 [J]. 工矿自动化, 2025, 51 (02): 100-105

备注/Memo:: 收稿日期：2025-11-18第一作者：郑勇(1986 —)，男，陕西渭南人，工程师，硕士，研究方向为煤炭开采技术研究工作。E-mail：zhengyong_2025@163.com

更新日期/Last Update: 1900-01-01