基于岩屑分形理论的岩石力学参数实时评价方法

现有的岩石力学参数获取方法依赖于钻井取心和测井资料,具有较大的时效限制。将分形几何理论应用于岩石力学参数实时预测的研究中。首先,利用完钻井深层岩心开展力学测试,建立岩石力学参数与多项测井物理量多元统计学模型。然后,开展上返岩屑颗粒的分形维数测试,并依据测井岩石力学模型计算出岩屑原位地层的岩石力学参数,将这些力学参数与岩屑分形维数建立统计学关系式。以准噶尔盆地玛湖凹陷Ma123井为例,通过测试上返岩屑的分形维数,进而实时计算出岩石的力学参数。完钻测井后基于测井反演模型绘制出该井岩石力学参数剖面。两种方法对比认为,基于岩屑分形维数计算的岩石力学参数具有较高的精度(误差小于10.9%)。同时,通过实践总结出一套基于录井岩屑分形理论的岩石力学参数的实时预测方法。该研究成果拓展了分形岩石力学的应用领域,为岩石力学参数的实时预测提供了新思路。

Real time prediction method of drilling rock mechanical parameters based on Cutting Fractal Theory

At present, the commonly used rock mechanics parameter acquisition methods rely on drilling coring and logging data, and have great lag and limitations. The fractal geometry theory is applied to the real-time prediction of rock mechanical parameters. First of all, the mechanical test is carried out by using the deep core after drilling, and the multivariate statistical model of rock mechanical parameters and multiple logging physical quantities is established. Then, carry out the fractal dimension test of rock debris particles, calculate the rock mechanical parameters of rock debris in-situ formation according to the well logging rock mechanical model, and establish a statistical relationship between these mechanical parameters and the fractal dimension of rock debris. Taking Ma123 in Mahu sag of Junggar Basin as the research object, the mechanical parameters of rock can be calculated in real time by testing the fractal dimension of the upward cuttings. After drilling and logging, the rock mechanical parameter profile of the well is drawn based on the logging inversion model. The comparison between the two methods shows that the rock mechanical parameters calculated based on the fractal dimension of rock debris have high accuracy (error less than 10.9%). At the same time, a set of real-time prediction methods of rock mechanical parameters based on the fractal theory of logging cuttings are summarized through practice. The research results expand the application field of fractal rock mechanics and provide a new idea for real-time prediction of rock mechanics parameters.