针对油藏地质中复杂地层岩性划分存在的模糊性和多解性 ,提出了一种基于单亲遗传算法的模式聚类划分复杂地层岩性的方法 .该方法通过隐含传统算法交叉算子功能的基因换位等遗传算子来实现进化操作 ,简化了遗传过程 ,同时避免了早熟收敛问题 .用PGA求解模式聚类问题可使聚类结果不依赖于初始聚类中心 .采用该方法对大庆油田10 8个地层样本进行处理 ,总符合率达 88.0 %.结果表明该方法可提高复杂岩性识别的精度 . Aiming at the fuzziness and multi-solution of complex stratigraphic lithology classification in reservoir geology, a method of pattern clustering based on single-parent genetic algorithm is proposed to divide complex formation lithology. By implicit traditional algorithm crossover operator function Genetic transposition and other genetic operators to achieve the evolutionary operation, simplifying the genetic process, while avoiding the premature convergence problem.Using PGA to solve the problem of pattern clustering can make the clustering results independent of the initial cluster center.Using this method, The results show that this method can improve the accuracy of complex lithology identification.