Song Kaoping

Neural Approach for Calculating Permeability of Porous Medium

Permeability is one of the most important properties of porous media. It is considerably difficult to calculate reservoir permeability precisely by using single well-logging response and simple formula because reservoir is of serious heterogeneity, and well-logging response curves are badly affected by many complicated factors underground. We propose a neural network method to calculate permeability of porous media. By improving the algorithm of the back-propagation neural network, convergence speed is enhanced and better results can be achieved. A four-layer back-propagation network is constructed to effectively calculate permeability from well log data. Spontaneous potential, resistivity of deep lateral log, resistivity of micro-gradient log, resistivity of micro-normal log, Interval transit time of acoustic log and resistivity of shallow lateral log are selected as the inputs, and permeability is selected as the output. There are 35 and 40 units used in the two hidden layers, respectively. During the training course, the correlation coefficient between the calculated permeability and the standard pattern is as high as 0.9937, the average absolute error between them is 0.046 μm2 and the average relative error is only 1.93%. For practical applications, the average relative error between the calculated permeability and actual permeability is also as low as about 10.0%.

Investigation on Mechanisms of Polymer Enhanced Oil Recovery by Nuclear Magnetic Resonance and Microscopic Theoretical Analysis

Polymer flooding is an efficient technique to enhance oil recovery over water flooding. There are lots of discussions regarding the mechanisms for polymer flooding enhancing oil recovery. The main focus is whether polymer flooding can increase sweep efficiency alone, or can increase both of sweep efficiency and displacement efficiency. We present a study on this problem. Oil displacement experiments on 4 natural cores show that polymer flooding can increase oil recovery efficiency by more than 12% over water. Moreover, photos are taken by the nuclear magnetic resonance (NMR) method both after water flooding and after polymer flooding, which show remaining oil saturation distribution at the middle cross section and the central longitudinal section. Analyses of these photos demonstrate that polymer flooding can increase both sweep efficiency and displacement efficiency.

Displacement Mechanism of Polymer Flooding by Molecular Tribology

Whether polymer flooding can enhance displacement efficiency or not is still a problem under debate. Laboratory experiment, numerical simulation and core data analysis are the commonly used means to study polymer flooding displacement efficiency. We discuss the limitations of these methods and employ molecular tribology to study the problem. The black?white ball action principle, i.e. the atom action model for describing the friction principle, is used to analyse the microscopic mechanism of oil displacement and describe the molecular interactions and displacement power during polymer flooding. Both tribology theory and dynamic rheological test show that molecular interactions during polymer flooding are bigger than that during water flooding. It is concluded that displacement efficiency of water flooding may be higher than that of polymer flooding at particular area; while polymer flooding can weaken the heterogeneity significantly, decrease ineffective injection and enhance the total displacement efficiency.

基于输入 K -近邻的正则化路径上SVR贝叶斯组合

在 e -不敏感支持向量回归( e -insensitive support vector regression, e -SVR)正则化路径的基础上, 提出基于输入 K -近邻的三步式SVR模型组合方法。在整个样本集上进行训练, 求得 e -SVR的正则化路径。由SVR正则化路径的分段线性性质确定初始模型集合, 并应用平均贝叶斯信息准则(Bayesian Information Criterion, BIC)策略对初始模型集合进行修剪以获得候选模型集合。该修剪策略可减小候选模型集合的规模, 提高模型组合的计算效率和预测性能。在预测或测试阶段, 根据样本输入向量采用 K -近邻法确定最终组合模型集合, 并实现贝叶斯组合预测。证明了 e -SVR模型组合的 L e -风险一致性, 给出了SVR模型组合基于样本的合理性解释。试验结果验证了正则化路径上基于输入 K -近邻的 e -SVR模型组合的有效性。

An Optimization Model for Stimulation of Oilfield at the Stage of High Water Content

This paper presents a method to predict pressure distribution and residual oil distribution for water flooding oilfield. To meet the demands of the integral design of oilfield development plan and the high-efficiency exploitation of oilfield, this paper proposes an optimization method for single stimulation measure and integral adjustment of a tract based on analysis of oilfield performance, economic evaluation and professional experience. This integral optimization method is helpful to decision making and optimize the adjustment project. Enforcement of the plan will help not only complete the mission of production, but can bring the maximum incomes under present conditions. So, this study is significant to improve the efficiency and effectiveness of field work.