改进MUSIC算法的超声波测风方法研究  

Research on ultrasonic wind measurement method with improved MUSIC algorithm

作　　者：唐心亮 宋欣朔 倪永婧[1] TANG Xinliang;SONG Xinshuo;NI Yongjing(School of Information Science and Engineering,Hebei University Science and Technology,Shijiazhuang 050018,China)

机构地区：[1]河北科技大学信息科学与工程学院,石家庄050018

出　　处：《重庆理工大学学报（自然科学）》2024年第4期283-289,共7页Journal of Chongqing University of Technology：Natural Science

基　　金：河北省高等学校科学技术重点研究项目(ZD2020318);河北省教育厅青年基金项目(QN2021066,QN2023185)。

摘　　要：针对传统超声波测风装置测风精度不高、抗噪声能力弱,提出了一种改进多重信号分类(multiple signal classification,MUSIC)算法的超声波测风方法。采用一种弧形6阵元超声波传感器阵列的测风结构,推导其阵列流型;在此基础上,添加小波阈值降噪算法提高信号信噪比,降低噪声信号协方差矩阵的秩;再使用PHAT加权广义互相关时延估计算法以提高时延估计的准确性,同时根据时延关系对传统MUSIC算法矢量矩阵进行改进;最后通过MUSIC算法实现对风速风向的测量。理论分析与仿真结果表明:改进后的MUSIC算法具有较好的抗噪性能和较高的风参数测量精度,测量风速绝对误差达到0.15 m/s,风向绝对误差达到2°,可以应用于对风参数要求较高的场景。Traditional ultrasonic wind measurement devices suffer low wind measurement accuracy and weak noise resistance.To address these issues,this paper proposes an improved MUSIC(multiple signal classification)algorithm for ultrasonic wind measurement.With the improvement,a wind measurement structure of an arc-shaped 6-element ultrasonic array is adopted,and its array popularity is derived.Based on this,a wavelet threshold denoising algorithm is employed to extract high signal-to-noise ratio ultrasonic emission signals to reduce the rank of the covariance matrix.The PHAT weighted generalized cross correlation time delay estimation algorithm is employed to improve the accuracy of time delay estimation.Finally,the MUSIC algorithm is adopted to measure wind speed and direction.Our theoretical analysis and simulation results show the improved MUSIC algorithm has excellent noise resistance performance and high wind parameter measurement accuracy.The measurement wind speed error is below 0.15 m/s,and the wind direction error is below 1°.Thus,it has huge potentials in scenarios with high requirements for wind parameters.

关 键 词：阵列信号处理 MUSIC算法 小波阈值降噪 广义互相关 风速风向测量 

分 类 号：TH765[机械工程—仪器科学与技术]