加权灰色线性回归组合模型在高铁隧道沉降监测中的应用

Application of Weighted Grey Linear Regression Model in Settlement Monitoring of High Speed Railway Tunnel

近年来,相关的研究与实践表明,变形监测的数据处理方法比较成熟,如回归模型、卡尔曼滤波模型、灰色模型、时间序列模型以及人工神经网络模型等各种模型,均经过了各种检验,而且有效地应用在变形监测技术中。然而单一的模型预测有其自身的局限性,因此,预测模型需要采用组合优化模型弥补单一模型的缺陷。本文主要阐述了加权灰色线性回归组合模型在高铁隧道沉降预测中的应用,通过与传统的GM(1,1)模型以及灰色线性回归组合模型进行对比。实验结果表明,加权灰色线性回归组合模型具有较高的预测精度。

In recent years, related research and practice have made the data processing methods of deformation monitoring become mature. For instance, Regression model、Kalman filtering model、Grey model、Time series model and Artificial neural network model and other models. And these models have been passed through a various of tests and used in the deformation monitoring technology in effectively. However, single model prediction has its own limitation. Therefore, the prediction model use combination optimization model to overcome the defects of single model. This paper mainly discusses the weighted grey linear regression combined model and its applied to the prediction of high speed railway tunnel subside. Then compared with the traditional GM（1,1） model and grey linear regression combined model. The experimental results show that the weighted grey linear regression combined model has higher prediction accuracy.