支持向量机在油田产量预测中的应用

Application of supportive vector machine to the prediction of production in oil fields

提出了一种基于支持向量机算法的油田系统建模理论,并且应用原—对偶算法来解决支持向量机中的二次规划问题.将支持向量机应用于预测油井产油量,预测实例表明,最大泛化相对误差为5.611%,预测值很接近油井的实际产量;与其它预测方法相比,该预测模型具有较高的预测精度.

A new way of modeling in oil fields based on the SVM is proposed in this paper.And the primal-dual algorithm is used to solve the dual programming in SVM algorithm.Supportive Vector Machine is used to predict the output of the oil wells in oil fields.The simulation example shows the most generalized relative error is 5.611%.The predicted value is near the actual output.The experiments have proved the new algorithm is superior to traditional modeling methods used in oil fields.