最优权组合预测法在路基沉降变形预测中的应用  

Application of optimal weight combination prediction method in subgrade settlement deformation prediction

作　　者：李文威 LI Wen-wei

机构地区：[1]中铁二院工程集团有限责任公司,四川成都610031

出　　处：《智能城市》2024年第2期97-100,共4页Intelligent City

摘　　要：文章以成熟的等间隔GM(1,1)灰色模型与BP神经网络模型为预测单模型,通过最优权算法分别计算出每个单模型的权重,构建组合预测模型。以某在建铁路工程路基沉降实测数据为例,分别建立了组合预测模型与两种单模型的预测结果比对。结果表明,在预测初期或末期,组合模型的预测精度各项指标均高于非等间隔灰色模型与BP神经网络模型,证明了最优权组合预测方法在路基沉降预测中的可行性与适用性。The article uses mature equidistant GM(1,1)grey model and BP neural network model as prediction single models,calculates the weight of each single model through the optimal weight algorithm,and constructs a combined prediction model.Finally,taking the actual measurement data of roadbed settlement in a railway project under construction as an example,a combination prediction model was established and the prediction results of two single models were compared.The experimental results show that the prediction accuracy of the combination model is higher than that of the non equidistant grey model and BP neural network model in both the early and late stages of prediction,demonstrating the feasibility and applicability of the optimal weight combination prediction method in predicting roadbed settlement.

关 键 词：路基沉降 非等间隔灰色模型 神经网络模型 最优权算法 

分 类 号：P228[天文地球—大地测量学与测量工程]