基于倾角导向梯度能量熵的断层检测方法

断层解释是三维地震数据精细构造解释的关键。三维不连续性地震几何属性分析技术的发展，为断层等不连续性地质边界的检测和识别提供了极大的便利。针对传统地震几何属性对随机噪声较为敏感，用于检测和识别微小断层效果不好这一现实问题，提出了一种基于倾角导向梯度能量熵的相干分析方法。首先利用梯度结构张量获取倾角信息，然后通过构建梯度相关矩阵计算局部地震数据的梯度能量熵值。高能量熵值代表数据的连续性很差，低能量熵值代表数据的连续性很好，因此可以利用能量熵值刻画三维地震数据的不连续性空间结构信息。理论模型和实际数据测试表明，新方法不仅抗噪能力较强，而且对断层等不连续性边界的检测效果比超级地震道第一代相干体（ST-C1）、超级地震道第三代相干体（ST-C3）和局部结构熵相干体（LSE）都好，可以有效地用于三维断层及其他地质边界的地震解释。

Fault identification based on dip-oriented gradient-energy-entropy coherence estimation

Locating small-scale strata discontinuities is one of the most challenging tasks,and these subtle geological features are significant since they are often associated with subsurface petroleum traps.Subtle faults,fractures,unconformities,reef textures,channel boundaries,thin-bed boundaries,and other structural and stratigraphic discontinuities have subtle geologic edges which may provide lateral variation on seismic data.Among different available geophysical techniques,3-D seismic discontinuity attributes are effective tools for highlighting discontinuities on seismic data.Conventional seismic discontinuity attributes are sensitive to noise and are not appropriate for detecting small-scale discontinuities.Thus,we propose a dip-oriented gradient-energy-entropy (DOGEE) coherence to detect subtle faults and structural features.The DOGEE coherence estimation method uses gradient structure tensor (GST) algorithm to obtain local dip information and construct a gradient correlation matrix to calculate gradient-energy-entropy.The proposed DOGEE coherence is robust to noise,and also improves the clarity of fault edges.It is effective for small-scale discontinuity characterization and interpretation.