基于熵值法的水平定向钻进地面沉降预测组合模型

A Combined Model for Predicting Ground Subsidence in HDD Based on Entropy Method

地面沉降是水平定向钻进施工必须面临的严重问题,在分析指数法、双曲线法以及三点法预测模型的基础上,引入信息学中信息熵的概念,采用熵值法综合以上模型的优点,对其赋权构成组合模型,并结合实际水平定向钻进施工案例对熵值法组合模型进行验证,结果表明该模型在整个工程监测过程中较其他单个模型预测更加精准,且满足工程要求。

Ground subsidence is a serious problem to be faced in Horizontal Directional Drilling( HDD) field at home and abroad. Based on the analysis of the three prediction models of index method,hyperbolic method as well as three-point method,this paper introduces the concept of the information entropy from informatics,and uses the entropy method to integrate the advantages of the three prediction models above and empower them to constitute a combined prediction model for greater precision. Then the paper uses the new combined prediction model and other three prediction models to fit the HDD case in Shanghai to compare their accuracy. The results show that the combined prediction model is consistent with the actual situation and more precise than other single prediction model throughout the project monitoring process,and it can also meet the engineering requirements.