论文部分内容阅读
水工程安全监测数据中不可避免地存在离群点,而应用最为广泛的最小二乘法(least square,LS)不具备剔除离群点的能力,反而更易吸收离群点,使回归曲线严重偏离实际。针对LS在此方面的缺陷,本文在最小化残差平方和理论的基础上,提出采用最小截平方和估计(least trimmed squares,LTS)方法来构建水工程安全监控模型。根据实际工程的监测资料并对监测资料分析处理,剔除离群点得到最优数据群。通过求解最优数据群的回归系数,进而得到最接近实际数据的拟合曲线。相比于LS估计,LTS估计所得结果更具有合理性、稳健性,且能够显著提高数据的预测精度。因此,LTS估计在水工程安全监测等数据分析中具有良好的应用前景。
However, the most widely used least square (LS) does not have the ability of removing outliers, but it is more likely to absorb outliers and make the regression curve seriously deviate from the actual . Aiming at the shortcomings of LS in this aspect, based on minimizing the sum of squares of residuals, this paper proposes to construct a water project safety monitoring model by using the method of least trimmed squares (LTS). According to the actual project monitoring data and monitoring data analysis and processing, remove the outlier to get the optimal data group. By solving the regression coefficient of the optimal data set, the fitting curve closest to the actual data is obtained. Compared with the LS estimation, the LTS estimation results are more reasonable and robust, and can significantly improve the prediction accuracy of the data. Therefore, LTS estimation has good application prospect in data analysis of water project safety monitoring.