基于稀疏采样的管道缺陷超声复合阵列检测机理研究

Massive testing data is a big problem for the existing pipeline ultrasonic inner testing, which has caused great difficulities in storing all the data,accomplishing real time processing, and rapidly detecting a variety of defects at one time. This project is aimed at researching the mechanism and method of detecting wall thickness changing flaws, leaks, axial and circumferential cracks by using composite array of ultrasonic transducers based on sparse sampling. We build the concept of ultrasonic special dictionary to achieve the sparsest indication of ultrasonic testing echo signal. The sparse sampling theory is introduced for the first time into sparse sampling mechanism and physical accomplishing method for the transducer composite array.We initiate rapid calculating method of characteristic parameters of flaw testing echo signals by sparse value, which makes it unnecessary to rebulild testing data, hence solving the problem of time-greedy. The composite array method for pipeline ultrasonic inner testing is employed to establish relevant model between array space structure parameters and characteristic parameters of flaw testing echo signal for flaw reconstruction with different space trend under the condition of sparse sampling.This research work will offer a sound scientific basis for the crux of pipeline ultrasonic inner testing.

管道超声内检测技术目前面临的突出问题是检测数据量太大，无法实现全部数据的存储和实时处理，无法同时实现多种缺陷快速检测。本课题拟开展基于稀疏采样的超声传感器复合阵列快速检测壁厚变化类缺陷、渗漏孔、轴向及周向裂纹的机理和方法研究。首次提出超声专业字典概念，实现超声检测回波信号最稀疏表达；创新性运用稀疏采样原理，提出超声传感器检测回波信号稀疏采样机理与前端物理实现方法；创新提出稀疏采样值快速解算缺陷回波信号特征参数方法，无需重建检测信号，解决稀疏采样值恢复原始检测信号的时间贪婪性问题；采用超声传感器复合阵列方法，建立阵列空间结构参数与缺陷回波特征参数关联模型，实现稀疏采样条件下不同空间趋向缺陷重构。本课题将为解决目前管道超声内检测技术难题提供科学依据。

以管道缺陷为检测对象，以超声波为检测手段，研究了基于稀疏采样的管道缺陷超声波复合阵列内检测技术，在准确保留超声检测信号特征参数信息的基础上，大幅度减少了检测过程中产生的超声回波信号采集数据量，便于全部检测数据的有效存储和实时处理。结合超声复合阵列传感器，能够快速检测出壁厚变化类、渗漏孔、轴向及周向裂纹多种类型的缺陷。通过本项目的研究，建立了专用的超声专业字典，实现了检测脉冲超声信号的最稀疏表达；在理论上建立了超声检测信号稀疏采样的理论框架，并且采用硬件电路首次实现了超声检测信号物理稀疏采样，突破了传统奈奎斯特采样的局限，电路性能满足设计要求；在不重构原检测信号的基础上，直接由稀疏采样得到的数据解算出了回波信号的声学特征参量，解决了稀疏采样值恢复原始检测信号的时间贪婪性问题；建立了超声传感器复合阵列空间结构参数与缺陷回波特征参数关联模型，实现了稀疏采样条件下不同类型缺陷重构与可视化成像。达到了项目申请时的既定目标。