基于多元回归方法的岩石物理性质预测模型研究

Study on the Prediction Model of Rock Physical Property Based on Multiple Regression Method

为了研究地层特性和地层层面划分,钻进过程中所钻地层的岩性识别必不可少。利用多元回归技术建立经验模型来预测岩石的物理性质,考虑了钻进过程中钻头直径、钻头转速、钻进速率和等效声级等参数,并用软件进行分析,采用t检验、F检验来验证模型的有效性,并导出预测模型的性能。结果表明,由回归模型计算出的岩石物理性质参数值(岩石抗压强度、纵波波速、岩石密度)与在实验室测得的值相当接近,模型的准确度分别为89.5%、96.7%、90.8%,所建立的模型能有效地用于完整岩石性质的预测。

In order to study strata characteristics and stratigraphic stratification, lithologic identification of the strata is essential during the drilling process. The physical mechanical property of rock mass can be predicted based on the empirical model, by considering the bit diameter, drilling speed, drilling rate and equivalent sound level parameters during the drilling process, based on software analysis method, the effectiveness of the model is verified by t-criterion and F-criterion, besides that, the performance of the prediction model is derived. The study results show that the rock physical mechanical parameters （rock compressive strength, longitudinal wave velocity and rock density） calculated by the prediction model are pretty close to the ones measured in laboratory, the prediction accuracy of the prediction model is 89.5%, 96.7% and 90.8%, which further show that the model can be used effectively to predict the properties of complete rock.