论文部分内容阅读
井层优选和施工方案优化是油气田重复压裂技术的核心 ,其“瓶颈”问题是“数据有限”和“模型与参数给不准”以及“许多问题的机理不清楚” ,无法获得问题的显式表达。文章尝试应用统计学来获取影响压裂效果的各项因素与压裂效果的关系模型和预测模型 ,从而优选施工井层和优化施工方案。实践证明 ,对于样本“数据有限”(小样本 )的情况下 ,支持向量机算法技术适应性强、精度高 ,在重复压裂研究领域中具有广阔的应用前景
Well site optimization and construction program optimization are the core of oil and gas field fracturing technology. The “bottleneck” problems are “limited data”, “no model and parameter” and “the mechanism of many problems are not clear” Expression. The article attempts to apply statistics to get the relationship between fracturing effect and various factors that affect the fracturing effect and predicts the model so as to optimize the construction well and optimize the construction scheme. Practice has proved that, for the sample “data limited” (small sample), the support vector machine algorithm has strong adaptability and high precision, and has broad application prospect in the field of repeated fracturing