基于Lasso算法的油田产量预测方法

随着油田的不断开采,油田的产量预测也变得越来越重要。目前有许多基于机器学习的预测方法,但大多数都不能给出具体的预测模型。本文提出了一种基于Lasso算法的预测方法,结合现场生产数据,选取一系列相关特征参数,通过对参数数据的初步分析,初步选取各个参数的函数形式,然后利用Lasso算法得到最终的预测模型,最终达到预测产量的目的。现场试验表明：该方法得到的预测模型比较准确,可解释性强,且预测精度高,可以应用于矿场产量预测。

A Prediction Method for Oilfield Production Based on Lasso Algorithm

As the continuous mining of oil fields, production forecast for oil fields is becoming more and more important. There are many prediction methods based on machine learning, but most of them cannot give a specific prediction model. In this paper, a prediction method based on Lasso algorithm is proposed. Combining the field production data, a series of related feature parameters are selected. Through the preliminary analysis of the sample data, the best function form of each parameter is selected, and then the prediction model is obtained by Lasso algorithm. Finally, the purpose of predicting production is achieved by the prediction model. The field test shows that the prediction model obtained by this method is accurate and is highly explanatory.Besizes,this method can get prediction results with high accuracy, and it can be applied to mine production forecasting.