储气库动态监测关键参数大数据预测模型研究

摘要

文章设计基于Hadoop大数据平台管理储气库动态数据,通过构造储气库动态监测解释方法,设计动态监测关键数据计算模型,利用深度学习模型预测动态监测关键参数,提高预测精度.通过对储气库地层中天然气物性参数计算,深度学习解释模型天然气物性参数精度达到97％、提高了储气库各储气层解释精度,可以满足储气库动态监测解释需求.%The construction of gas storage has been largely carried out,and it is imminent that the dynamic data of monitoring the gas storage.This paper design is based on gas storage dynamic data of Hadoop data platform management,by constructing dynamic monitoring interpretation method of gas storage,to design the calculation model for key data of dynamic monitoring,and by using the deep learning model to predict the key parameters of dynamic monitoring,to improve prediction accuracy.Through the calculation of gas storage formation in natural gas physical parameters,accuracy of physical parameters of natural gas of the deep learning interpretation model has reached 97％ and interpretation accuracy of gas storage.Every gas bearing reservoir has been improved which can meet the needs of gas storage dynamic monitoring interpretation.