基于层控蚁群层位自动追踪的Wheeler转换方法及其应用

对地震数据进行预处理,增强同相轴的连续性,然后引入支持向量机技术对地震剖面上的点进行分类处理,同时以地震振幅、瞬时相位、层位倾角、信息素浓度等信息作为蚁群层位追踪的评价函数,提出改进蚁群搜索的地震层位对比和自动追踪算法。在此基础上,将时深域沉积旋回韵律体对比追踪结果转换到Wheeler域,实现时间域和Wheeler域的同步解释,从而有效揭示沉积组合的空间关系,突出缺失和叠加模式,使在年代地层域中提取时间切片成为可能,对沉积、侵蚀和沉积时间有更深刻的理解。实际地震资料的应用结果验证了方法的有效性。

Wheeler conversion method based on layer-controlled horizon automatic tracking with ant-colony algorithm and its application

Continuity of the same phase axis is enhanced with pretreatment of seismic data. Then, the support vector machine technology is introduced to classify the points on the seismic profile. An improved ant-colony algorithm for seismic horizon correlation and automatic tracking algorithm is proposed, in which the information of seismic amplitude, instantaneous phase, dip angle, and pheromone concentration is used as the evaluation function of ant-colony horizon tracking. On this basis, contrast and tracking results of sedimentary cycle rhythm in time-depth domain are transformed to Wheeler domain which realizes synchronous interpretation of time domain and Wheeler domain. Thus, the spatial relationship of sedimentary assemblages is revealed effectively with highlight loss and overlay mode to make it possible to extract time slices in the age-strata domain, which have a deeper understanding with the time of deposition, erosion and deposition. The validity of the method is verified by the application of the actual seismic data.