基于SBF-ISVD的带式输送机声信号增强方法  

Acoustic signal enhancement method of belt conveyor based on SBF-ISVD

作　　者：张晓东 张玉强 杜方鹏 马波[2,3] 游卿华 ZHANG Xiaodong;ZHANG Yuqiang;DU Fangpeng;MA Bo;YOU Qinghua(CHN ENERGY Yulin Energy Co.,Ltd.,Yulin 719000,China;College of Mechanical and Electrical Engineering,Beijing University of Chemical Technology,Beijing 100029,China;Beijing Key Laboratory of High-End Mechanical Equipment Health Monitoring and Self Recovery,Beijing University of Chemical Technology,Beijing 100029,China)

机构地区：[1]国能榆林能源有限责任公司,陕西榆林719000 [2]北京化工大学机电工程学院,北京100029 [3]北京化工大学高端机械装备健康监控与自愈化北京市重点实验室,北京100029

出　　处：《机电工程》2024年第10期1875-1884,共10页Journal of Mechanical & Electrical Engineering

基　　金：青龙寺煤矿智能矿山建设关键技术研究项目(GJNY-22-132)。

摘　　要：在带式输送机的声学诊断过程中,其声信号受混响及背景噪声的干扰十分严重,为此,通过分析混响产生的原因和声信号的组成等,提出了一种基于超指向性波束形成(SBF)去混响和改进奇异值分解(ISVD)降噪的声信号增强方法。首先,利用基于能量变化最大值进行了最优频带选择,确定了包含故障信息较多的频带;然后,利用SBF去除了混响对声信号的干扰,采用ISVD方法对去混响后的信号进行了降噪处理,并对信号进行了包络谱分析,对比了实际测得的故障特征频率和理论的故障特征频率,对带式输送机的故障特征进行了提取;最后,设计了实验,采集了实验数据,利用该方法对煤矿现场采集到的数据进行了分析验证,并将其与加权预测误差算法(WPE)和线性约束最小方差(LCMV)相结合的方法以及递归最小二乘法(RLS)进行了对比。研究结果表明:与原信号相比,经SBF-ISVD方法处理后,实验数据包络谱中内圈故障特征频率153.1 Hz及其倍频312.5 Hz处的幅值明显提高,信噪比从-31.39 dB显著提高至-25.4 dB。基于SBF-ISVD的声信号增强方法去混响和降噪效果显著,轴承故障特征提取效果较好,可实现复杂环境噪声下带式输送机声信号增强的目的。In order to solve the problem that the sound signal of belt conveyor is seriously disturbed by reverberation and background noise in the acoustic diagnosis,the reasons of reverberation and the composition of sound signal were analyzed,an acoustic signal enhancement method based on super directional beamforming(SBF)and improved singular value decomposition(ISVD)was proposed.Firstly,the optimal frequency band was selected based on the maximum energy variation to determine the frequency band with more fault information.Then,SBF was used to remove the interference of reverberation on the sound signal,ISVD method was used to reduce the noise of the signal after reverberation,and the envelope spectrum of the signal was analyzed.The fault characteristic frequency measured in practice was compared with the theoretical fault characteristic frequency,and the fault characteristics of the belt conveyor were extracted.Finally,the test was designed,the data were collected,and the data collected from the coal mine site were analyzed and verified by using this method.The method was compared with the weighted prediction error algorithm(WPE),the linear constrained minimum variance(LCMV)and the recursive least square method(RLS).The research results show that:comparing with the original signal,the amplitude at the inner circle fault characteristic frequency of 153.1 Hz and the frequency doubling of 312.5 Hz in the envelope spectrum of test data processed by SBFISVD method is significantly improved,and the signal-to-noise ratio is significantly increased from-31.39 dB to-25.4 dB.The sound signal enhancement method based on SBF-ISVD has remarkable effect on removing reverberation and reducing noise,and the bearing fault feature extraction effect is good,which can realize the sound signal enhancement of belt conveyor under complex ambient noise.

关 键 词：皮带输送机 轴承故障诊断 声学诊断 混响消除 降噪效果 频带 超指向性波束形成 改进奇异值分解 

分 类 号：TH133.33[机械工程—机械制造及自动化] TD528.1[矿业工程—矿山机电]