应用时序分析ARIMA模型预测抽油井生产系统效率

Application of ARIMA Model in Time Series Analysisto Predict the Efficiency of Pumping Well Production System

抽油井生产系统效率是抽油系统的重要工况指标。系统效率预测是指通过对历史数据的分析和研究,找出系统效率内部变化规律及影响因素之间的关联,然后对系统效率做出预测。以前,油田多采用机理模型实现系统效率预测,但因抽油井系统效率的不确定性、复杂性和时间性的特点,预测效果并不理想。因此本文建立时序分析中的差分自回归移动平均模型对华北油田抽油井生产数据进行了拟合及抽油机井生产系统效率预测。此数据模型着重强调了相邻时间点的抽油井系统效率的相关性。实验结果表明持续更新最新数据点的ARIMA模型能够及时纠正预测方向,大幅提高抽油井生产系统效率预测的准确度。

The efficiency prediction of oil well production system is an important index of pumping system. System efficiency prediction is based on the relationship between the internal variation and the influencing factors, which is learned from the analysis of historical data. Previously, oil fields used mechanism model to predict system efficiency. However, the outcome is not desired due to the uncertainty, complexity and timing of pumping well system efficiency. Therefore, this paper establishes the Autoregressive Integrated Moving Average Model(ARIMA) in time series analysis to predict and analyze the efficiency of Huabei oil field pumping well production system. This model emphasis the effects of the previous nearby data on the prediction by keep updating the latest data. The result indicates ARIMA model could adjust the direction of the prediction and hence improve the accuracy of the prediction for oil production system efficiency.