基于时频掩码和稀疏表示的带输送机托辊故障诊断  

Fault Diagnosis of Belt ConveyorRoller Based on Time Frequency Mask and Sparse Representation

作　　者：杨筱彧 佘影 罗明华 方崇全[1,2] YANG Xiaoyu;SHE Ying;LUO Minghua;FANG Chongquan(Chongqing Research Institute,China Coal Technology and Engineering Group,Chongqing 400039,China;State Key Laboratory of The Gas Disaster Detecting,Preventing and Emergency Controlling,Chongqing 400037,China)

机构地区：[1]中煤科工集团重庆研究院有限公司,重庆400039 [2]瓦斯灾害监控与应急技术国家重点实验室,重庆400037

出　　处：《自动化与仪器仪表》2025年第1期43-46,51,共5页Automation & Instrumentation

基　　金：中煤科工集团重庆研究院有限公司重点研发项目(2022ZDXM01)。

摘　　要：带输送机作为煤矿主要的输送设备,其平稳健康运行对于煤矿的安全生产至关重要。严峻的工作环境使得带输送机托辊容易出现故障,造成严重的安全隐患,因此,对托辊的故障诊断具有重要的意义。针对井下存在非平稳噪声干扰和信号频率分布稀疏的问题,提出了一种基于时频掩码和稀疏表示的托辊故障诊断方法。首先,根据声音信号的时频图提取出每个信号振动周期的标准模态,并以标准模态度量原始时频图得到时频掩码矩阵。通过时频掩码去除非平稳噪声的干扰,防止故障误报。然后,利用稀疏表示中K奇异值分解(K-Singular Value Decomposition,K-SVD)算法学习得到健康样本的稀疏完备字典,并采用该字典分别对测试样本进行重构,根据重构误差实现故障诊断。实验结果表明,提出的时频掩码技术可以有效地滤除非平稳噪声;稀疏表示具有很高的故障诊断精度,并且可以提高诊断效率。As the main conveying equipment in coal mines,the smooth and healthy operation of belt conveyors is of great significance for the safety production of coal mines.Due to the harsh working environment,belt conveyor idlers are prone to malfunctions,posing serious safety hazards.Therefore,fault diagnosis of idlers is of great significance.In response to the problems of non-stationary noise and sparse signal frequency distribution underground,this paper proposes a fault diagnosis method for idlers based on timefrequency mask and sparse representation.Extract the standard mode of each signal vibration period from the time-frequency map of the voiceprint signal,and measure the original time-frequency map using the standard mode to obtain the time-frequency mask matrix.Remove non-stationary noise interference through time-frequency masking to prevent false alarms of faults.Utilize the K-SVD algorithm in sparse representation to learn a sparse complete dictionary of healthy samples,and use this dictionary to reconstruct the test samples separately,achieving fault diagnosis based on reconstruction errors.The experimental results show that the time-frequency mask technique proposed in this paper can effectively filter out non-stationary noise;Sparse representation can effectively improve the efficiency of fault diagnosis and has high fault diagnosis accuracy.

关 键 词：托辊故障诊断 非平稳噪声 时频掩码 稀疏表示 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]