基于Burg最大熵法的管道腐蚀超声检测

Ultrasonic Inspection of Pipeline Corrosion Based on Burg Maximum Entropy Method

超声波检测是输油管道在线内检测的重要方法之一,但在管道壁厚测量应用中却经常出现大量的误差,甚至得出相反的结果。本文进行了较深入的超声检测回波分析,指出了由回波信号的前两个脉冲的时间间隔计算管道剩余壁厚的检测方法将产生较高的误检率和漏检率,提出了一种基于Burg最大熵谱估计的信号处理检测方法,Burg法按最大熵原则外推自相关函数,递推计算模型预测误差滤波器反射系数,避免了其他自回归功率谱估计方法中自相关函数的计算。短时间序列、高分辨、抗干扰能力强的功率谱估计算法是超声管道内检测的关键技术。实验结果表明该方法检测精度高、谱估计分辨率高,适用于输油管道在线管壁腐蚀检测。

The ultrasonic inspection is one of the important ways to inspect the wall-loss defects on-line for the oil pipeline. However, many errors or opposite results are obtained in practical application. Aiming at this problem, an effective method based on Burg maximum entropy power spectral estimation is put forward. Based on the maximum entropy principle of the extraplated self-correlation, the Burg method can recurrently calculate the reflect coefficient of prediction error filter （PEF） model, thus avoiding the calculation of self correlation function. Power spectral estimation characterized by short time serial, higher resolution and anti-noise is a key technology to inspect the inner wall of pipeline with ultrasonic. Experimental results show that the method increases the spectral resolution and the detection accuracy, and is suitable for on-line detection of pipeline corrosion and other wall-loss defects.