论文部分内容阅读
将最优分割,边缘检测与多种群遗传算法相结合,设计了一种适用于多参数层序地层自动划分的边缘最优智能划分算法.该算法综合利用测井参数的Fisher比与表征测点奇异性的Lipschitz指数(Lipschitz Exponent,LE)构造优化指标;以钻孔参数测点为基因,以有序测点组为染色体,通过参数控制实现不同种群同时进行优化搜索,采用移民算子沟通各种群协同进化;实现综合利用多参数测井数据求取最优化地层分界线.在河北平原第四纪地层岩性划分中的实际应用,表明该方法的自动分层结果符合地质实际,计算速度快,分层效率高,对河北平原山前冲积洪积、中部冲积湖积、东部冲积海积等不同沉积类型地层的划分工作都有较好适用性.
Combining optimal segmentation and edge detection with multi-population genetic algorithm, an edge optimal intelligent partitioning algorithm is proposed for automatic classification of multi-parameter sequence stratigraphy.Based on Fisher’s ratio of logging parameters and characterization of measuring points The Lipschitz exponent (LE) of the singularity is used to construct the optimization index. Taking the measuring points of the drilling parameters as the genes and the ordered measuring points as the chromosomes, different populations are optimized for searching simultaneously by the parameter control, and the immigration operator is used to communicate The symbiotic evolution of the population and the realization of comprehensive utilization of multi-parameter logging data to obtain the optimal stratigraphic boundary.The practical application of Quaternary stratigraphic lithology classification in the Hebei Plain shows that the automatic stratification result of the method accords with the geological reality and the calculation speed Fast and high stratification efficiency. It has good applicability to the division of alluvial-pluvial sediments, alluvial-lacustrine deposits and alluvial deposits in the east of Hebei Plain.