储层岩石压缩系数的一种预测方法

A prediction method for reservoir rock compressibility

岩石压缩系数是油藏工程研究中的一个重要参数,它标志着油气开采过程中岩石所能提供的能量大小。实际应用时,如果按照经验公式求取岩石压缩系数,往往会产生较大误差。为此,引入近年来预测效果较好的支持向量回归机技术,以压力、孔隙度为输入参数,建立支持向量回归机预测模型,用来对未知地层岩石压缩系数进行预测。实际应用表明,预测结果与岩石实际压缩系数之间的相对误差基本都控制在允许范围之内,可见该方法的预测效果很好,可行性高,从而为油气储层岩石压缩系数的预测求取开拓了一种新思路。

Rock compressibility is a very important parameter in the study of reservoir engineering. It indicates that how much energy the rock can provide during the oil and gas exploitation. In practical application, if we calculate the rock compressibility according to the empirical formula, the result will cause major error. Therefore, this paper introduces a method of supporting vector regression, which has preferable forecasting effect to set up the forecasting model for rock compressibility of the unknown reservoir. In this model, we take pressure and porosity as input to calculate rock compressibility. The result indicates that the fractional error between the prediction result and the real value is very small, so this method has preferable effect and feasibility, which affords a new clue of calculating rock compressibility for oil and gas reservoir.