HHT在探地雷达检测路基质量中的应用

Application of Hilbert-Huang transform in detecting the quality of roadbed by ground penetrating radar

为探明路基中隐藏病害的类型和空间位置,应用探地雷达方法对公路路基质量进行了无损检测.在地层介质强衰减和噪声干扰作用下,探地雷达信号存在信噪比低、深部复杂目标体难以识别的问题.针对探地雷达路基实测信号非线性和非平稳的特点,采用信号均衡和希尔伯特-黄变换相结合的方法,首先对数据进行均衡处理以突出深层弱信号,在此基础上对单道信号进行经验模态分解以获得固有模态函数,通过希尔伯特变换对固有模态函数的瞬态属性特征进行分析,进而将希尔伯特-黄变换应用于整个探地雷达的数据剖面,最后通过综合对比瞬时振幅剖面和瞬时相位剖面对探测结果进行了分析和解释.结果表明,信号均衡可以有效缓解深部信号衰减问题,基于希尔伯特-黄变换可以从多属性参数角度对探地雷达信号进行处理与解释,可以较好的识别异常体的位置和属性信息,提高了探地雷达解释结果的可靠性.

In order to detect the types and spatial locations of the hidden diseases in roadbed, the Ground Penetrating Radar(GPR) technology was applied to evaluate the quality of highway subgrade nondestructively. Under the effects of strong attenuation of formation media and noise interference, the signal to noise ratio of GPR signal is low and the complex target body in deep part is difficult to be identifies in the data profile. The method combining signal equalization and Hilbert-Huang Transform(HHT) was adopted to deal with the nonlinear and non-stationary characteristics of the measured roadbed GPR signals. First, the data was equalized to highlight weak signal from the deep reflection interface and then the empirical mode decomposition of a single channel signal was carried out to obtain the intrinsic mode functions, and then the transient properties of intrinsic mode functions are analyzed by Hilbert transform, and then the Hilbert-Huang transform is applied to the whole GPR profile. At last, the detection results are analyzed and interpreted by synthetically comparing the instantaneous amplitude profile and the instantaneous phase profile. The results show that, signal equalization can effectively alleviate the deep signal attenuation problem, based on the Hilbert-Huang transform the GPR signal can be processed and analyzed from multiple attribute parameters, and the position of the anomalous body can be well recognized and improve the reliability of ground penetrating radar interpretation results.