论文部分内容阅读
对岩土三轴试验数据采用两种线性回归法计算所得的抗剪强度参数c、φ值会有差别。为了揭示这两种回归方法差别的根本原因,深入最小二乘法线性回归的基本原理指出,按大小主应力回归的结果是最佳估值,而按p-q曲线回归的计算结果不可靠,为了得到统一的结果,对后者所基于的最小二乘法进行了修正;推导了修正后的回归系数计算公式,证明了修正后两种方法得到的结果相同,并用一组千枚岩三轴试验数据进行了验证。结果表明,修正后用这两种方法确定三轴试验抗剪强度参数值都将得到同样的最佳无偏估值。
There are differences between the shear strength parameters c and φ calculated by two linear regression methods on the triaxial test data of geotechnical. In order to reveal the root cause of the difference between the two regression methods, the basic principle of in-depth least-squares linear regression shows that the result of regression by size principal stress is the best estimate and the result of regression by pq curve is unreliable. The least square method based on the latter is modified; the modified regression coefficient calculation formula is deduced, and the results obtained by the two methods after the correction are proved to be the same. A set of phyllite triaxial test data is used verification. The results show that the modified two methods to determine the triaxial test shear strength parameter values will be given the same best unbiased estimates.