论文部分内容阅读
按I.M.Gelland等人建立的模式识别Cora-3修改算法研究华东地区的地质构造、地貌和地球物理特征并用于识别本地区破坏性地震。破坏性地震易发生在新构造运动的沉降区内、白垩纪或新生代构造(断陷)盆地的内部或边缘、新构造垂直差异运动的地质块体边界附近以及活动断层数较多的地方。 “投票”结果表明误识率较低(ε=0.061—0.098)。根据“投票”结果,划分出五个D节点相对密集的地震危险区。 控制试验表明识别结果是比较稳定和可靠的。
According to the pattern recognition Cora-3 modified algorithm established by I.M. Gelland et al., The geological structure, geomorphology and geophysical characteristics of East China are studied and used to identify the devastating earthquakes in this area. Destructive earthquakes easily occur in the subsidence area of the Neotectonic movement, inside or edge of the Cretaceous or Cenozoic tectonic (rift) basin, near the boundaries of newly formed tectonic blocks with vertically divergent motions, and where the number of active faults is large. Voting results showed a lower false-positive rate (ε = 0.061-0.098). According to the results of “voting”, five D seismic hazard zones with relatively dense nodes are classified. Control tests show that the recognition results are relatively stable and reliable.