基于模糊变步长LMS的超声测厚信号时延估计方法
DOI:
CSTR:
作者:
作者单位:

1.蚌埠市特种设备监督检验中心;2.合肥工业大学

作者简介:

通讯作者:

中图分类号:

TB92;TB551;O426.9

基金项目:

安徽省市场监督管理局科技计划项目(2022MK051)


Time delay estimation method of ultrasonic thickness measurement signal based on fuzzy variable step size LMS
Author:
Affiliation:

Bengbu Special Equipment Supervision and Inspection Center

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    准确获取飞行时间(Time of Flight,TOF)对于高精度超声测厚至关重要。针对传统超声信号时延估计中固定步长最小均方(Least Mean Square,LMS)算法在收敛速度与稳态误差之间存在固有矛盾,以及现有变步长算法依赖瞬时误差和固定函数模型导致在非平稳回波信号下适应性差的问题,提出一种基于模糊变步长LMS的超声测厚信号时延估计方法。分析超声回波信号的时变特性,摒弃单一误差反馈机制,提取输出信号与期望信号的局部相关系数误差及其变化量作为模糊控制器的双输入特征。设计零阶Sugeno模糊推理系统,建立输入特征与步长因子之间的非线性映射规则,实现步长因子的自适应动态调节。利用模拟回波信号进行不同信噪比下的仿真测试,结果表明:相较定步长LMS算法、双曲正切函数变步长LMS算法以及基于瞬时误差的模糊变步长LMS算法,所提方法的综合性能更优,在保证快速收敛的同时显著降低了稳态失调误差,具有更高的测量精度和抗噪性能。搭建超声测厚实验平台,以标准量块为对象进行测厚实验,结果表明:所提方法对不同厚度量块的测量相对误差均小于其他三种LMS算法,最大相对误差为0.71 %。模糊变步长LMS时延估计方法可为高精度超声TOF计算提供可行方案选择和技术支撑,有利于促进超声无损检测技术发展,具有一定的工程应用价值。

    Abstract:

    Accurate acquisition of time of flight (TOF) is essential for high-precision ultrasonic thickness measurement. A time delay estimation method for ultrasonic thickness measurement signal based on fuzzy variable-step least mean square (LMS) was proposed, addressing the issues that the inherent contradiction between the convergence speed and the steady-state error of the fixed-step LMS algorithm in the traditional ultrasonic signal time delay estimation, and the poor adaptability of the existing variable-step algorithm under non-stationary echo signals due to the dependence on the fixed function model. The time-varying characteristics of ultrasonic echo signal were analyzed, and the single error feedback mechanism was abandoned. The local correlation coefficient error and its variation between the output signal and the desired signal were extracted as the dual input characteristics of the fuzzy controller. The zero-order Sugeno fuzzy inference system was designed, and the nonlinear mapping rule between the input feature and the step size factor was established to realize the adaptive dynamic adjustment of the step size factor. The simulated echo signals were used to carry out simulation tests under different signal-to-noise ratios. The results show that compared with the fixed step size LMS algorithm, the hyperbolic tangent function variable step size LMS algorithm and the fuzzy variable step size LMS algorithm based on instantaneous error, the comprehensive performance of the proposed method is better. The steady-state offset error is significantly reduced while ensuring rapid convergence, with higher measurement accuracy and anti-noise performance. The experimental platform of ultrasonic thickness measurement was built, and the thickness measurement experiments were performed on gauge blocks. The results show that the relative errors of the proposed method for different thickness gauge blocks are all smaller than that of the other three LMS algorithms, and the maximum relative error is 0.71 %. The fuzzy variable step size LMS time delay estimation method can provide feasible scheme selection and technical support for high-precision ultrasonic TOF calculation, which is conducive to promoting the development of ultrasonic nondestructive testing technology and has certain engineering application value.

    参考文献
    相似文献
    引证文献
引用本文
分享
相关视频

文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2025-11-20
  • 最后修改日期:2026-01-20
  • 录用日期:2026-02-05
  • 在线发布日期:
  • 出版日期:
文章二维码