油气田（井）生产动态预测的灰色预测模型

本文针对辽河油田荣３７块天然气藏（井）的生产动态预测中，生产数据的离散状态及不均衡的实际，较完整、系统地介绍了所引入的几个使用效果好，拟合精度高的一般灰色系统ＧＭ（１，１）模型及为解决油气田（井）生产过程中因关井、生产间断、或因工程上的与地质上的原因生产出现随机性急剧变化而引入两个改进的ＧＭ（１，１）模型－非等间距和阶跃模型。通过几种模型特征的分析对比及在辽河油田荣３７块全气藏及气井的动态拟合预测

GRAY FORECASITING MODEL IN DYNAMIC PRODUCTION FORECASTING OF OIL

In this paper, we introduced systematically and completely several effective and precise common gray system GM(1,1) models in accordance with the divergent and unbalanced production data in dynamic forecasting of production in 37 Rong natural gas wells of Liaohe oil fields (wells). Two improved GM(1,1) models,non equidistance model and step spring model,were also introduced to solve the problems of random and rapid diversification in production owing to closed in production,interrupted production or engineering and geological reasons. By analyzing and comparing the characters of these models and utilizing them in the dynamic forecasting of 37 Rong natural gas wells of Liaohe oil fields, we proved that these models were convenient, rapid, effective and pecise. It seemed that the two improved GM(1,1) models were especially satisfactory in resolving the problem of lack of data due tostoppage of production and dealing with the non equidistanced serial data caused by sudden variation of production or the data which jumped and descended steeply. Therefore, we hereby provided several new gray forecasting models for the dynamic forecasting of production.These models remedied the deficiency caused by sole method application and inefficiency of past forecasting.