基于支持向量机的蒸汽吞吐井调剖参数优化

由于调剖措施效果受多种因素的影响,它们之间是一种极其复杂的非线性关系，常规方法很难在调剖措施和影响因素之间建立一种确定的关系。在对蒸汽吞吐井调剖影响因素分析的基础上，利用支持向量机的方法，以周期增油量为评价指标，建立了调剖效果和各个影响因素之间的关系模型，开发了调剖注入参数优化软件，对蒸汽吞吐井最佳的调剖剂注入量和周期增油量进行了预测。结果表明，建立的预测模型样本学习的误差在5%以下，模型具有较好的预测效果和实用价值。

OPTIMIZATION OF PROFILE PARAMETERS IN STEAM HUFF AND PUFF WELL BASED ON SUPPORT VECTOR MACHINE

ecause the effect of profile adjusting is influenced by various factors, and the relationship between them is very complex and nonlinear, it is hard to establish the relationship between the profile adjusting and factors using traditional means. According to the affecting factors of profile adjusting in steam huff and puff well, a relation model between the profile adjusting effect and factors is established based on support vector machine. The computer software to optimize huff and puff parameter of profile adjusting in steam huff and puff well is developed. Using the model, the optimal intake volume of profile adjusting agents and increased oil production in one circle is predicted. The predicted results show that the inspection error of the model is controlled within 5% and the model is feasible and applicable.