模拟退火PSO优化算法在油水两相流产出剖面解释中的应用  

Application of Stimulated Annealing-particle Swarm Optimization Algorithm to Production Profile Interpretation for Oil-water Two Phase Flow

作　　者：张闪[1] 郭海敏[1] 时新磊[1] 

机构地区：[1]油气资源与勘探技术教育部重点实验室长江大学地球物理与石油资源学院,湖北荆州434023

出　　处：《测井技术》2010年第3期289-292,共4页Well Logging Technology

基　　金：国家科技重大专项项目资助(2008ZC05020-005)

摘　　要：提出了用模拟退火POS优化算法进行油水两相流产出剖面解释的方法。建立了含等式约束和不等式约束的优化目标函数,导出了流体流量、流体平均密度和持水率的理论响应方程。借助优化算法求出非相关函数达到最小值的解。该算法综合了粒子群优化算法和模拟退火优化算法的特点,在计算过程中自动生成迭代初始值,减少人为误差;不但接受好的解,还以一定的概率接受差的解,保证搜索过程中求出全局最优解。编写了产出剖面优化处理程序。对文×井的实际数据处理结果表明该方法效果良好。Put forward is a stimulated annealing-particle swarm optimization algorithm for production profile interpretation of oil-water two phase flow. Established is the optimization objective function including equality constraints and inequality constraints. Derived are the theoretical equations about fluid flow rate, fluid average density and water holdup. The solution of incoherence function when it reaches minimum value is calculated by optimization algorithm. This algorithm combines the feature of stimulated annealing algorithm and particle swarm optimization algorithm. The stimulated annealing-particle swarm optimization algorithm generates iteration initial value automatically which reduces personal error. And it ensures the result it seeks to is the global optimal solution by accepting good solutions as well as a certain proportion of bad solutions. The optimization procedures are programmed and its practical application in Wen × well shows good result.

关 键 词：测井解释 两相流 产出剖面 模拟退火算法 粒子群优化算法 

分 类 号：P631.413[天文地球—地质矿产勘探]