基于改进DNA算法的地震层位拾取技术

Seismic horizon picking technology based on improved DNA algorithms

针对常规地震DNA算法难以设置正确的搜索基因,并且拾取出的地震层位可能出现过于散乱、连续性不强的现象等问题,在保证搜索基因高效合理的前提下,提出复式五点采样的方法建立基因库,减少人为干预。应用基于互相关的二级窗式算法判断相邻两基因片段是否来自同一层位,并用三次样条插值连接同一层位,使拾取出的层位具有更好的连续性和实用性。结果表明:本方法拾取出的层位效率更高,同相轴更完整,减少了人为误差;由于本方法是全局识别,较之传统地震同相轴追踪技术,易于推广到三维数据。

For conventional seismic DNA algorithms,it is difficult to set the correct search gene,which results in scattered seismic horizons which lack strong spatial continuity.Under the premise of ensuring an efficient and reasonable search gene,this paper proposes a multiplex five-point sampling approach to establish the gene bank,which reduces subjective human intervention.Based on the cross-correlation of a two-level window algorithm,this application first determines whether two adjacent gene fragments come from the same horizon;a cubic spline interpolation is then used to connect the same horizon.Subsequently the picked horizon has better continuity and practicability.The results show that this method is more efficient and in-phase axis is more complete.The improved method reduces human errors,and is easily transferable to three-dimensional data compared with traditional seismic event tracking technology.