基于组合模型的高铁隧道围岩收敛变形预测

Prediction of surrounding rock convergence deformation of high-speed railway tunnel based on combined model

为提高围岩收敛变形预测精度,依托阳山高速铁路隧道,基于贝叶斯参数优化的经验模型预测收敛变形的趋势项,然后采用支持向量回归算法修正经验模型的预测结果,据此构建基于组合模型的高铁隧道围岩收敛变形预测模型.将均方误差和平均绝对百分误差作为评价指标,与同类型支持向量机回归模型的预测结果进行对比.结果表明,经验模型的预测值与目标值随时间变化的趋势一致,但对于数据中的波动部分,预测值与目标值存在较大偏差.误差修正模型较好地预测数据中的波动部分,对提高预测精度效果明显.在2组实测数据中,组合模型预测值的均方误差相较于经验模型降幅分别达到97.0%和93.4%,且较同类型支持向量机回归模型具有更高的预测精度.

To improve the prediction accuracy of tunnel convergence deformation,taking the Yangshan high-speed railway tunnel as an example,the empirical model(EM)based on Bayesian parameter optimization was used to predict the trend term of tunnel convergence deformation.Then,the support vector regression(SVR)algorithm was applied to correct the prediction results of EM.A prediction model combining EM and SVR was established.The root mean square error(RMSE)and the mean absolute percentage error were taken as the evaluation indexes,and the results of the prediction model were compared with those of the same type of support vector regression models.The results show that the predicted values of EM are consistent with the trend of the target values over time.However,there is a big deviation between the predicted values and the target values for the fluctuating part of the data.The error correction model can accurately predict the fluctuation part of the data and improve the prediction accuracy obviously.In the two sets of measured data,the RMSE of the predicted values of the combined model decreases by 97.0%and 93.4%,respectively,compared with EM,and the prediction accuracy is higher than those of the same type of the support vector machine regression models.