摘要
灰色聚类方法是一种新型的多参数非线性分类法，能大大降低仅由单一参数评价所带来的片面性，并能克服线性方法的多解性，可以用来对裂缝发育程度之类的复杂地质参数进行更准确的评价。通过提取包括测井参数、录井显示特征以及沉积微相等多达１５种控制或反映裂缝发育程度的参数，研究各种参数随裂缝发育程度的变化规律，建立各个评价参数对于裂缝发育程度评价级别的白化函数。以已测试层作为建模样本层，对已测试层进行模型试算以确定模型是否可以用来进行预测，用所建立的白化函数与未测试层的评价原始参数进行运算，对未知层的裂缝发育程度作出预测，预测结果发现有一定数量的未测试层裂缝发育程度评价级别较高，对其开发潜力应予以重视。研究表明该方法参数提取适当、运算法则科学、建模精度较高、预测结果可信。
As a novel multiparameter non-linear classification, the grey cluster method may be used for more accurately evaluating certain complicated geological parameters as the degree of fracture development, etc., by which the one-sidedness formed by the evaluation only with single parameter can be greatly decreased and the multi-solvability caused by use of linear methods can be overcome. On the basis of extracting 15 parameters as log parameter, logging shows and sedimentary microfacies, etc., the change law of each parameter with the degree of fracture development was investigated and a manifest function expressing the assessment grade of each parameter to the degree of fracture development was set up. According to the data collected from tested layers, a pilot calculation was carried out to determine whether the model may be used to make a prediction, then, on the basis of above-mentioned manifest function and of the initial parameters of evaluating untested layers, the degree of fracture development in the untested layers was predicted by the model. Through prediction it is found that the assessment grade of the degree of fracture development in certain untested layers is high, therefore their development potential should be taken into account. The research results indicate that this method is of many advantages, such as suitable evaluation parameters, scientific operational rule, high modeling accuracy and credible prediction outcome, etc.
出处
《天然气工业》
EI
CAS
CSCD
北大核心
2001年第6期70-73,共4页
Natural Gas Industry
关键词
气藏
单井裂缝发育程度
灰色聚类
评价参数
白化函数
预测
Fracture development, Degree, Grey cluster, Evaluation parameter, Manefest function, Prediction
