大型环枝状油田注水管网系统运行优化

针对大型环枝状油田注水管网系统考虑实际运行工艺约束、流量、压力约束等多个约束条件,以管网运行能耗与注水井注入量偏差损失成本总和最小为目标建立混合整数非线性规划(MINLP)模型,模型贴合管网实际运行工况。针对大规模复杂MINLP模型的求解,提出蚁群算法结合线性近似与单纯形法的混合求解策略。结果表明:所建立的数学模型考虑管网实际运行约束较为全面,求解对象可以为大型环枝状复杂结构的注水管网。所提出的求解策略能够求解规模较大、变量之间关系复杂的MINLP模型,在保证求解效率的前提下,计算结果最优性较强,较大程度上提高了该注水系统运行的稳定性与经济性,满足现场工艺要求。
As for the large-scale looped and branched waterflooding pipeline network system of oilfields, the practical operation process, flowrate, and pressure constraints are taken into account. A mixed integer nonlinear programming (MINLP) model was established, taking the least cost sum of the pipeline network operation consumption and the waterflooding volume deviation loss of waterflooding wells as the object.The model can conform to the practical operation conditions. To solve the large-scale complex MINLP model, the ant colony optimization was proposed, combined with a hybrid solution strategy of linear approximation and simplex method. The results show that, this established model is enough to consider the practical operation constraints of the pipeline networks. It is applicable to solve the large-scale looped and branched waterflooding pipeline network system. The proposed strategy can be used to solve the MINLP with large scale and complex variables. To satisfy the requirement of solving efficiency, the computational results tend to the optimal ones. It has greatly improved the stability and economy of this waterflooding system, which satisfies the requirement of technology for the oil field