如今大地电磁测深技术(MT)在油气勘查中的应用越来越广泛,但MT数据因具有非线性、非平稳和非最小相位的特征,不符合以傅里叶变换为基础的传统谱分析的基本要求.最新发展起来的Hilbert-Huang(HHT)是处理非线性、非平稳信号的有效方法.文中从三个方面讨论了HHT在MT数据处理中的应用:1提出利用Hilbert谱对MT数据进行时段筛选,提高信号品质;2利用经验模态分解和Hilbert谱分析MT数据中的噪声分布特征,进行噪声压制;③利用经HHT处理后的数据从Hilbert谱统计估算阻抗张量,使MT数据非平稳性带来的估算偏差最小化.实测数据处理的结果表明:HHT方法处理MT信号是有效的,在抑制噪声、提取信号中有用信息方面优于传统方法,基于HHT谱的阻抗估计消除了传统功率谱方法中大地电磁响应函数出现个别频点分散、误差棒较大的现象,使地质参数的估计精度和资料的可解释性得到明显提高.%Nowadays the need for using Magnetotelluric (MT) sounding in oil and gas exploration is growing. While MT data series are nonlinear, non-minimum phase and typical non-stationary random signals, they do not meet the basic requirements of conventional methods which are based on the Fourier transform. The latest developed Hilbert-Huang Transform (HHT) is an effective method for processing of nonlinear and non-stationary random signals. In this paper application of HHT to Magnetotelluric Data Processing were discussed in the following three aspects: (1) Using Hilbert time-frequency energy spectrum to select MT signal sessions is helpful to improve signal quality; (2) Using empirical mode decomposition method and Hilbert spectrum can effectively analyze the noise distribution of MT data so the noise can be suppressed; (3) By using processed HHT data, the impedance tensor was estimated from the Hilbert spectrum statistics and the estimation error which was caused by non-stationary characteristics of MT data was minimized. Results show that the HHT method is effective to process MT data, and it is a better method than traditional methods in noise suppression and extraction of useful information from signals. By applying HHT method, MT response curve effectively avoids irregular expansion of certain frequencies with big error bars or shape distortions, that ensures the following qualitative analysis and quantitative interpretation.
展开▼
