基于IOWHA算子的路基沉降加权组合预测方法研究  

Study on weighted combination model based on IOWHA operator in prediction of subgrade settlement

作　　者：赵亚红 王金星 张丽华 周文国 郝延锦 ZHAO Yahong;WANG Jinxing;ZHANG Lihua;ZHOU Wenguo;HAO Yanjin(Architectural Engineering College,North China Institute of Science and Technology,Beijing 101601,China;Beijing Digsur Science and Technology Co.,Ltd,Beijing 100012,China)

机构地区：[1]华北科技学院建筑工程学院,北京101601 [2]北京帝测科技股份有限公司,北京100012

出　　处：《铁道科学与工程学报》2018年第10期2529-2534,共6页Journal of Railway Science and Engineering

基　　金：国家自然科学基金资助项目(51178185);廊坊市科技支撑计划资助项目(2016011014)

摘　　要：针对传统的加权组合预测模型赋权过程中的缺陷,分析诱导有序加权调和平均算子(IOWHA)的优势,建立一种基于诱导有序加权调和平均算子的泊松-指数曲线组合预测模型。其基本思想是,以单一模型在每个时刻的预测精度作为诱导值,根据其大小排序、赋权值,以组合预测误差倒数平方和最小为准则,得到最优模型的最优权系数解,进行组合模型预测。结合工程实测数据,利用组合模型进行预测。研究结果表明:基于IOWHA算子的组合预测模型预测平均绝对误差及平均绝对百分比误差均小于单一模型以及传统的最优加权几何平均组合模型,预测结果能很好的反映沉降趋势。Aiming at the defects of the traditional weighted combination forecasting method,the characteristics of the induced ordered weighted harmonic averaging operator(IOWHA)was analyzed,and a Poisson exponential curve combination prediction model based on the induced ordered weighted harmonic averaging operator was established.The basic idea is that the prediction accuracy of a single model at each moment was used as the induction value,according to its size sorting;the data of model was assigned by weighted.In the context the quadratic sum of the reciprocal of combinatorial prediction error was minimum,the optimal weight coefficient was obtained by the optimal model so that combined model was applied to predict.The verification was carried out through the section settlement data.The results show that the mean absolute error and the mean absolute percentile error of the combined prediction model based on IOWHA operator are smaller than the single model and the traditional optimal weighted geometric mean combination model.The prediction results can settlement trend of the settlement very well.

关 键 词：IOWHA 泊松.指数曲线 路基沉降 加权组合预测 

分 类 号：U416[交通运输工程—道路与铁道工程]