SSA-Elman神经网络模型在建筑物沉降预测中的应用  

Application of SSA-Elman Neural Network Model in Building Settlement Prediction

作　　者：兰丽景 陈晓婷 毛洪孝 LAN Lijing;CHEN Xiaoting;MAO Hongxiao(Zhejiang Zhenbang Geographic Information Technology Co.,Ltd.,Quzhou 324000,China)

机构地区：[1]浙江振邦地理信息科技有限公司,浙江衢州324000

出　　处：《测绘与空间地理信息》2024年第4期203-206,共4页Geomatics & Spatial Information Technology

摘　　要：为了提高建筑物沉降变形预测精度,最大限度地减少监测数据中非变形噪声分量对预测结果的影响,本文在Elman神经网络模型的基础上引入奇异谱分析方法,构建新的SSA-Elman神经网络模型。首先利用SSA方法提取沉降监测数据中的趋势分量与周期分量,剔除噪声分量,提高监测数据信噪比;其次通过Elman神经网络模型分别对趋势分量、周期分量进行预测,得到对应分量预测结果;最后重构趋势分量与周期分量预测结果得到最终预测结果。通过实测建筑物沉降数据分别对Elman神经网络模型与SSA-Elman神经网络模型进行建模与预测,结果表明,SSA-Elman神经网络模型的预测精度更高,更适应长周期预测。In order to improve the prediction accuracy of building settlement deformation and minimize the impact of non deformation noise component in monitoring data on prediction results,in this paper,singular spectrum analysis is introduced based on Elman neural network model(SSA,single spectrum analysis)method to construct a new SSA-Elman neural network model.Firstly,the SSA method is used to extract the trend component and periodic component in the settlement monitoring data,eliminate the noise component and improve the signal-to-noise ratio of the monitoring data;secondly,the Elman neural network model is used to predict the trend component and periodic component respectively,and the corresponding component prediction results are obtained;finally the final prediction result is obtained from the prediction results of structural trend component and periodic component.Through the measured building settlement data,Elman neural network model and SSA-Elman neural network model are modeled and predicted respectively.The results show that SSA-Elman neural network model has higher prediction accuracy and is more suitable for long-term prediction.

关 键 词：Elman神经网络模型 奇异谱分析 建筑物 沉降预测 去噪 

分 类 号：P25[天文地球—测绘科学与技术] TB22[天文地球—大地测量学与测量工程]