针对直接互相关被动时延估计法定位管道异常振动事件存在噪声干扰影响定位精度的问题,提出了基于三阶累积量及自适应滤波时延估计的管道异常事件定位方法.该方法对顺、反两路异常振动信号进行三阶自累积量和互累积量估计,抑制高斯相关噪声和对称分布噪声.然后利用自适应滤波时延估计算法对三阶自累积量和互累积量信号的时延进行迭代计算,在不依赖先验知识的情况下抑制非高斯相关噪声.经现场实验证明,该方法可以准确地对管道异常事件进行定位,对噪声具有很好的抑制作用,改善了直接互相关时延估计的性能.相对于直接互相关时延估计方法,相对定位误差由2.7％降低到0.6％,定位一致性提高了三倍,平均定位精度可达14 m.%Aiming at the problem that in the direct cross-correlation method for time delay estimation of pipeline abnormal vibration events, passive positioning is easily affected by noise, a new method based on third-order cumulant and adaptive filter delay estimation for locating the pipeline vibration was put forward. The abnormal vibration signals were processed using third-order self-and cross-cumulant estimation, so the related Gaussian noise and symmetrical distribution noise can be suppressed. An adaptive filter was adopted to estimate the time delay of third-order self-cumulation and crosscumulation , which can suppress the non-Gaussian correlated noise even without the prior knowledge. The field test proves that, the method can accurately locate the pipeline abnormal vibration events, suppress the noise commendably, and improve the performance of time delay estimation. Comparing with the direct cross-correlation time delay estimation method, the relative positioning error reduces from 2. 7% to 0. 6% , positioning consistency increases three times, and the average positioning accuracy is up to 14m.
展开▼
