路基沉降的加权组合预测方法研究
Research on weighted combination prediction method for subgrade settlement

基于路基沉降实测数据对沉降进行预测是一种常用的施工质量管理方法,然而单一预测模型往往可靠度不高。对此,通过系统分析当前常用的3种路基沉降预测模型,在此基础之上,对这3种方法的预测结果赋予不同的权值,进而加权求和得出最终沉降值,旨在减小预测结果的随机性,提高预测精度。研究结果表明:预测数据可靠度要优于单一模型,对比类似组合算法,在保证计算精度的基础上,计算效率也有了显著提升。
To predict the subgrade settlement basing on the original settlement data is a common method for construction quality management. However, the reliability of a single prediction models are often not high. Therefore, this paper analyzed the current three common subgrade settlement prediction model systematicallyand given different weights for the results of these three methods. The final settlement prediction value can be achieved. It is aimed to improve forecast accuracy. In the prediction test, the results show that this method is superior to a single model. Compared with similar combination algorithm, the accuracy can be ensured, but computational efficiency has been significantly improved