基于小波去噪的自回归滑动平均模型在地铁隧道地表沉降预报中的应用

Application of ARMA Model Based on Wavelet Denoising to Ground Subsidence Prediction of Subway Tunnel

利用小波分析对地表沉降监测数据进行去噪处理,再使用传统的自回归滑动平均模型（auto-regressive and moving average model,ARMA）对去噪后的序列进行建模预报。以青岛市地铁3号线某条观测线实测数据为例,分别采用传统ARMA模型和基于小波去噪的ARMA模型进行了预报对比分析,结果表明基于小波去噪的ARMA模型取得了较高的预报精度。

Traditional ARMA（auto-regressive and moving average）model does not preprocess the observation data during predicting the ground subsidence of subway tunnel,which makes the settlement observations contain a lot of noise and does not help to reflect the real settlement law and modeling forecast.In this paper,wavelet analysis method is used to denoise the surface subsidence monitoring data,the ARMA model is used for modeling and forecasting.Taking the measured data of an observation line of Qingdao Metro Line 3 as an example,the traditional ARMA model and the ARMA model based on wavelet denoising are applied to compare and analyze the results.Results show that the ARMA model based on wavelet denoising has achieved a high forecast accuracy.