基于压缩感知和宽带俞式低通整形滤波器的地震低频信息特征分析与补偿

碳酸盐岩储集层已成为世界石油新发现储量的重要组成部分,识别该类储层对地震数据的信噪比、分辨率以及成像精度提出了更高的要求.本文从地震低频信号缺失的问题出发,首先研究了低频信号缺失对子波、合成地震记录和波阻抗反演的影响,其次分析了深层碳酸盐岩裂缝储层中弱信号低频缺失的特征.针对低频信号缺失问题,本文利用压缩感知理论,并结合反射系数的稀疏特性,提出了自适应计算L1范数权重因子的方法,同时构建了改进的宽带俞式低通整形滤波器,在不影响地震高频信号的同时对地震弱信号进行低频补偿.结果表明,缺失低频信号,会使子波旁瓣变大,合成记录出现假同相轴,厚层波阻抗反演畸变,深层碳酸盐岩裂缝储层弱信号难以识别;而本文方法有效地补偿了深层碳酸盐岩裂缝储层弱信号10Hz以下的频率成分,使得波组反射特征更加清晰,深层弱信号成像质量得到改善,为进一步有效识别深层碳酸盐岩裂缝储层建立了基础.

Feature analysis and compensation of seismic low-frequency based on compressed sensing and broad-band Yu-type low-passing shaping filter

The carbonate reservoir has become a significant portion of new-added reserve for oil industry. But the identification of carbonate reservoir requires higher SNR, resolution and imaging precision for seismic data. Starting from the problem of low-frequency seismic signal loss, this paper first studies its influence on wavelet, synthetic seismic records and wave impedance inversion. The features of weak signal from deep carbonate fracture reservoir in the absence of low-frequency component is also analyzed. To solve the problem, an adaptive method for computing L₁ norm weighting factor is proposed with the use of compressed sensing theory and the sparse property of reflection coefficient. The paper also establishes an improved broad-band Yu-type low-passing shaping filter, which compensates the low-frequency components of seismic signal and preserves the high-frequency information at the same time. The results show that the missing of low-frequency signal can lead to the increase of wavelet side-lobe, emergence of false events in the synthetic records, and the distortion of thick layer wave impedance inversion, thus making it difficult to identify the weak signal from deep carbonate rock fracture reservoir. The method presented in this paper can effectively compensate frequency component below 10Hz of weak signal from deep carbonate fracture reservoir, after which the reflection events group becomes clearer and the weak signal imaging quality is improved. This further provides good basis for the identification of deep carbonate fracture reservoir.