摘要
针对气测解释的随机性和模糊性的特点,提出一种两阶段模糊聚类算法。该算法通过引入密度参数对最大最小距离算法作了改进,以改进后的最大最小距离算法对数据集进行粗聚类,再以粗聚类所得的聚类中心为初始聚类中心执行标准模糊C-均值算法,得到类中心以及各数据类别。用于某油田某区块的储层油气性识别的实践表明,该算法实现简单、准确率较高、稳定性好,优于标准FCM算法。
Because of the random and fuzziness of gas logging data interpretation, a two-stage clustering algorithm is proposed. In this algorithm the density is applied to improve the Maxmin algorithm and get the cluster centers through the improved Maxmin algorithm. Then the cluster centers are used as the initial cluster centers of FCM algorithm to get the final cluster centers and the kind of the data. The algorithm is used in the identification of oil-bearing reservoirs and gas-beating reservoirs. The results show that this algorithm is simple and highly accurate and more stable than the standard FCM algorithm.
出处
《计算机工程与设计》
CSCD
北大核心
2009年第4期1027-1029,共3页
Computer Engineering and Design
基金
中石油青年创新基金项目(04E7015)
关键词
气测
最大最小距离算法
密度参数
模糊聚类
模糊C-均值算法
gas logging
maxmin algorithm
density parameter
fuzzy clustering
fuzzy c-means clustering algorithm
