| 杨志浩.基于BP神经网络的底水油藏控水压裂选段新方法[J].地质与勘探,2017,53(4):818-824 |
| 基于BP神经网络的底水油藏控水压裂选段新方法 |
| A new method for choice of water-control fracturing segments in horizontal wells based on the BP neural network system |
| 投稿时间：2017-03-13 修订日期：2017-05-02 |
| DOI： |
| 中文关键词: 控水压裂段选择 神经网络 底水脊进模型 水平井 |
| 英文关键词: water-control fracturing section choice, neural network, bottom water coning model, horizontal well |
| 基金项目: |
|
| 摘要点击次数: 1515 |
| 全文下载次数: 798 |
| 中文摘要: |
| 随着水平井压裂技术的普遍应用，压裂段的选择成为影响增产效果的主要因素之一，但针对底水油气藏，在压裂时同时考虑控水同样重要。影响底水油气藏压裂效果的因素很多，因素之间具有高度不确定的非线性关系，给控水压裂选段分析带来了诸多不便。利用传统的单因素回归等方法，显然不能满足底水油藏多种影响因素下水平压裂段的选择。针对此问题，通过油藏工程理论和现场经验，确定了避水高度、地层各向异性、天然裂缝发育程度、地应力差值、水平井曲直偏差、储层与水层之间隔层发育程度六个参数作为影响控水压裂段选择的主要因素，建立定性分析方法。通过底水脊进模型确定了参数界限，结合BP神经网络系统的原理，建立定量分析。利用计算机语言编制底水油气藏水平井压裂段控水选择软件，通过实际应用，表明其对现场施工具有一定的指导作用。 |
| 英文摘要: |
| With the widespread application of horizontal well fracturing technology, how to choose fracturing segments has become one of the primary factors affecting production increase. Especially for bottom water reservoirs, how to avoid premature of bottom water in fracturing production is particularly important. There are many factors that influence the fracturing effect of bottom water reservoirs, with a complex nonlinear relationship between the various factors,which has led to many inconveniences in water fracturing selection. The use of traditional single factor regression and other methods, obviously cannot meet the choice of horizontal fracturing sections of bottom water reservoirs under a variety of factors.To solve this problem, applying reservoir engineering theory and field experiences,we identified the height of water avoidance,formation anisotropy,natural fracture development degrees,crustal stress difference,deviation of horizontal well,stratosphere development level between reservoirs and aquifers as main factors affecting the fracturing sections selected, and then established qualitative analysis.In terms of the model of bottom water coning,we determined the parameters of the boundaries.Combining with the BP neural network theory,we developed a quantitative analysis.Using computer technology we compiled water-control choice software for horizontal well fracture sections of bottom water and gas reservoirs. Practical application of this new approach shows that it can play a certain guidance role in site operation. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
