Abstract: Currently, the classification criterion of reserves are determined by the scope of the values of criteria such as geological attributes, reservoir phydical parameters and etc., which require all attribute values of one block should be just right in the existing range of criteria, otherwise it would be difficult to divide the hard-to-recover reserves into different categories. To solve this problem, this paper combines with Fuzzy c-Means clustering algorithm(FCM)and combination weighting approach to classify hard-to-recover reserves. First, FCM is used to automatically search for the optimal category number of reserves based on effect indexes. Then, a combination weighting model is established based on the minimal error-sum of deviation of subjective weights and deviation of objective weights, which is used to compute the weights of attributes and the values of effect indexes. Finally, the categories that blocks belonge to is judged according to the result of FCM. To verify the validity of model, this paper applies it to the classification problem of hard-to-recover reserves from an oil field in the 10th Oil Production Plant of PetroChina Daqing Oilfield LLC, which would conduct the rolling development of hard-to-recover reserves.

Key words: reserves classification, fuzzy-means(FCM), combination weighting approach, hard-to-recover reserves

摘要： 目前储量的分类标准是通过划分指标值的范围来确定的,这就要求所有指标值恰好符合既定的指标范围,否则难以划分储量类别。为克服这一问题,本文结合模糊c-均值算法和组合赋权法实现难采储量的分类。首先基于效益指标运用模糊c-均值算法自动搜索储量的最佳类别,再利用主客观赋权偏差最小的思想,构建组合赋权模型,确定属性指标的权重,并计算储量效益指标值,结合模糊c-均值结果判别难采储量类别。最后以大庆某油田为实例,对其难采储量进行了分类,有效指导难采储量滚动开发决策。

关键词: 储量分类, 模糊c-均值, 组合赋权, 难采储量