时间推移试井技术研究及应用  

TIME EXTRAPOLATION WELL TEST TECHNIQUE AND APPLICATIONS

作　　者：屈怀林[1] 曾晓慧[2] 黄波[2] 郭大立[2] 

机构地区：[1]西南石油大学研究生院,四川成都610500 [2]西南石油大学石油工程学院,四川成都610500

出　　处：《西南石油学院学报》2006年第3期20-22,共3页Journal of Southwest Petroleum Institute

基　　金："油气藏地质及开发工程"国家重点实验室项目(石油汁算技术专项研究)成果之一

摘　　要：针对压力监测井点的动态监测信息在时间上的不连续性和历史的继承性,根据历次的试井资料、生产动态信息并结合地质静态资料,应用时间推移试井理论,采用神经网络技术,结合模拟退火算法,实现了对静压、有效渗透率和表皮系数等试井参数变化趋势的预测。通过实例分析,说明了利用试井历史信息进行时间推移预测的有效性。Aim at the dynamic monitor information that the pressure monitor well order on time of not the consecution and histories inherit according to trying the well data and producing the dynamic information and combining the geology static data of many times in the past, applying time extrapolation well test theories and the nerve network technique, combining to the emulation backs the fire calculate way, carrying out to static pressure,effective penneate rate and epidermis factor to try the estimate that the well parameter changes . Finally, applied example analysis, explain to make use of appling the well test history information to carry on time extrapolation well test predicting is useful.

关 键 词：时间推移试井 神经网络 模拟退火 

分 类 号：TE353[石油与天然气工程—油气田开发工程]