摘要
为了编制油田的效益规划,实现对利润、产量、成本、投资多个目标的合理决策,根据油田开发的实际情况,将油田的构成细分为9个部分,建立了利润、产量、成本和投资的非线性多目标优化模型,将利润、产量、成本、投资4种油田开发指标分成不同的优先级进行优化。采用ARMAX模型预测模型参数,用多项式拟合描述产量与工作量之间的非线性关系,在对多目标设置不同优先级的基础上,应用基于遗传算法的逐层非线性多目标算法,较快地得到最优解,对油田效益规划的编制具有重要意义。
For arranging a profit-centered oil field programme, by which multiple objects, profit, production, cost and investment, could be optimized, this thesis classifies oil fields in 9 sections according to practice on oil field development, as well establishes a non-linear multiple-objects optimized model, taking a priority level based consideration for the above mentioned multiple objects. AMARX predict model is used for parameters; multinomial fitting is used for describing non-linear relationship between production and workforce. After setting different priority level to objects, descendiblity non-linear order layer multiple objects arithmetic was used to get optimized solution with little time consuming. Techniques provided here are important to arranging a profit-centered programme for oilfields.
出处
《石油钻探技术》
CAS
2003年第4期59-61,共3页
Petroleum Drilling Techniques
基金
胜利油田"十五"重大科技攻关项目(编号10528)