【摘要】 断层属性体是自动解释断层的基础,但这些属性也包括来自地层成层性和各种噪声产生的假象,这些假象使得断层位置和断层系统交切关系变得模糊不清,很难准确识别和拾取断层。与噪声不同的是,断层线的延伸均有一定的方向。基于此特点,本文通过计算数据块的梯度方向约束蚂蚁追踪的范围,根据计算数据块的方向一致性区分断层和噪音,用蚁群算法在相干数据体或其他断层增强数据体中自动识别和追踪断层。此外,使用Zhang-Suen算法对蚁群追踪到的断层数据体中断层线进行细化处理,得到可以应用于地震解释的断层线。通过数值模型,验证了该方法在自动追踪断层和抑制影响断层识别的线性噪声中应用的可行性和有效性。实际地震资料的断层自动提取的结果表明,基于梯度方向约束的群算法是一种有效的断层追踪方法,在提高计算效率、减小计算量和抑制线性噪声方面具有一定优越性。更多还原

【Abstract】 The attribute of faults is the basis on fault automatic tracking.However,the attributes include some false information from stratification and noises,which make the position and intersection relationship of fault system not clear,and it is difficult to identify and take up the fault.Furthermore,the fault extension has a certain direction,which is different from noise.Based on this characteristics,the gradient direction of the block is calculated to constrain the ant tracking range,and the consistency of direction is calculated to differentiate between the fault and the noise.Finally,faults will be tracked automatically in the coherence data volume or other fault enhanced data volume by ant colony algorithm.In addition,automatic tracking faults are refined by Zhang-Suen algorithm to obtain the fault line that can be applied to seismic interpretation.The results of numerical model show that this method is feasible and available in the automatic tracking fault and suppression of noise.Field data processing results also show that algorithm based on constraints of the gradient direction is an effective method to trace the fault,and has certain advantages in improving calculation efficiency,decreasing calculation amount and reducing noise.更多还原