煤炭工程 ›› 2025, Vol. 57 ›› Issue (4): 123-130.doi: 10. 11799/ ce202504018

• 研究探讨 • 上一篇    下一篇

基于振动信号的带式输送机齿轮箱故障诊断研究

胡敬,袁龙江   

  1. 国家能源集团神东大柳塔煤矿
  • 收稿日期:2024-04-03 修回日期:2024-09-11 出版日期:2025-04-10 发布日期:2025-05-28
  • 通讯作者: 胡敬 E-mail:a1362509809@163.com

Fault diagnosis of belt conveyor gear box based on vibration signal

  • Received:2024-04-03 Revised:2024-09-11 Online:2025-04-10 Published:2025-05-28
  • Contact: Jing Hu E-mail:a1362509809@163.com

摘要:

在煤矿开采过程中带式输送机一旦发生故障将会影响煤矿的正常开采,给企业造成巨大的经济损失甚至是人员伤亡。因此,建立基于振动信号的带式输送机齿轮箱故障诊断模型十分必要。本研究以带式输送机振动信号为研究对象,分析典型故障类型,使用基于峭度-排列熵评价准测的变分模态分解对振动信号进行降噪,建立基于最小二乘支持向量机的带式输送机故障诊断模型,并使用改进粒子群优化算法针对故障诊断模型的惩罚系数与核参数进行优化,最终得到故障诊断的结果,故障诊断模型准确率达到95%以上,相比于传统SVM、RF、BPNN等故障诊断方法准确率提高6%,实现故障精确诊断,提高故障处理效率,提高带式输送机运行的可靠度,保证煤矿的安全生产。

关键词:

带式输送机齿轮箱 , 信号处理 , 故障诊断 , 变分模态分解 , 最小二乘支持向量机

Abstract:

In the realm of coal mining, the belt conveyor emerges as a pivotal mechanized equipment, playing a crucial role in the transportation of coal. Its operational integrity is instrumental in maintaining both the production efficacy and safety standards of coal mining activities. The gearbox, a core component in the belt conveyor system, significantly influences the conveyor's reliability and overall operational performance. Any malfunction within this apparatus during coal mining can lead to substantial disruptions, potentially incurring severe economic setbacks and jeopardizing worker safety. Thus, the formulation of a sophisticated fault diagnosis model for the belt conveyor gearbox, predicated on vibration signal analysis, is imperative. This study focuses on the vibration signals emitted by the belt conveyor as a basis for investigation. It delves into the identification of characteristic fault types, employing variational mode decomposition enhanced by kurtosis-permutation entropy evaluation for signal denoising. A novel diagnostic approach is proposed through the development of a fault diagnosis model based on the Least Square Support Vector Machine (LSSVM), incorporating an optimization of normalized and kernel parameters via an advanced particle swarm optimization algorithm. The efficacy of this model is evidenced by its diagnostic accuracy rate surpassing 95%, a significant improvement of at least 6% over conventional diagnostic methodologies. This leap in diagnostic precision not only facilitates swift and efficient fault rectification but also elevates the reliability of belt conveyor operations, thereby safeguarding the integrity of coal mine production processes.

中图分类号: