混沌粒子群算法优化最小二乘支持向量机的混凝土强度预测  

Prediction of concrete strength based on least square support vector machine optimized by chaotic particle swarm optimization

作　　者：张静[1,2] 刘向东[1] 

机构地区：[1]内蒙古工业大学材料科学与工程学院,呼和浩特010051 [2]内蒙古建筑职业技术学院机电与暖通工程分院,呼和浩特010070

出　　处：《吉林大学学报（工学版）》2016年第4期1097-1102,共6页Journal of Jilin University:Engineering and Technology Edition

基　　金：国家自然科学基金项目(51164023);国家教育部新世纪优秀人才支持计划项目(NCET-13-0847);内蒙古自治区教育厅项目(NJZC13344)

摘　　要：为了获得更理想的混凝土强度预测结果,提出一种混沌粒子群算法优化最小二乘支持向量机(LSSVM)的混凝土强度预测模型。首先采集混凝土强度数据,并进行归一化处理。然后采用LSSVM对混凝土强度与影响因子之间的变化关系进行建模,并采用混沌粒子群算法搜索最优LSSVM参数。最后采用具体混凝土强度预测实例对其性能进行分析。结果表明,本文模型可以准确描述混凝土强度与影响因子间的变化关系,提高了混凝土强度预测精度,具有一定的实际应用价值。In order to obtain precise prediction of concrete strength,aprediction model of concrete strength is proposed,which is based on Least Square Support Vector Machine（LSSVM）optimized by Chaotic Particle Swarm Optimization（CPSO）.First,the data of concrete strength are collected and normalized.Second,LSSVM is used to model the relationship between the concrete strength and the impact factors,and CPSO algorithm is applied to search the optimal parameters of LSSVM.Finally,the prediction performance is analyzed on the concrete strength data.Results show that the proposed model can accurately describe the relationship between the concrete strength and the impact factors,and improve the prediction accuracy of the concrete strength.This model has certain practical value.

关 键 词：土木工程 混凝土强度 最小二乘支持向量机 混沌粒子群算法 通用性测试 

分 类 号：TU528[建筑科学—建筑技术科学]