基于最优权重联合模型的滑坡位移预测研究

摘要

A comprehensive combination model for predicting landslide displacement is proposed in this context,using the recorded time-series deformation data of Qinglong landslide along Shang-Rui expressway.The elementary displacement prediction is made by the Support Vector Machine (SVM) model,the Exponential Smoothing model,and the GM (1,1) model.To comprehensively evaluate the results by these models,we combine the results of these models and introduce the accuracy matrix into the evaluation work in order to get optimal weight of each model.A rational prediction result can be attained based on the so-called combination model.The comparison between the prediction results and in-situ measurement shows that the prediction precision of the proposed combination model is satisfactory and it is much higher than the individual models.Therefore,the proposed combination model is feasible and effective on the landslide prediction.%在对联合预测模型的权重研究基础上,依托沪瑞高速公路贵州段晴隆滑坡的监测资料,首先建立了该滑坡位移预测的3个非线性预测模型:支持向量机模型、指数平滑模型、GM(1,1)模型;根据各模型训练结果,然后引入准确性矩阵,得到各单模型的权重,建立多模型联合预测模型,解决多模型预测结果取舍和可靠性评判问题.将联合模型与各单一模型的预测结果对比分析得出:联合模型的预测精度高于任何单一模型的预测精度.因此,运用最优权重联合模型能有效地提高滑坡预测预报精度.