混凝土强度预测的两种非线性模型比较研究  

Strength prediction of high strength concrete using two nonlinear methods

作　　者：朱学兵[1] 

机构地区：[1]宁波工程学院建筑工程学院,浙江宁波315000

出　　处：《混凝土》2011年第12期28-30,共3页Concrete

摘　　要：混凝土的强度的预测是一个复杂的问题,受多种因素的影响。采用两种先进的非线性算法支持向量机与小波支持向量机,建立了混凝土强度预测的两种非线性预测方法。研究结果表明:两种方法的预测结果与实测结果吻合较好,小波支持向量机的预测精度较支持向量机精度高,在混凝土的强度预测中具有较好的适应性。The concrete strength under influence of several factors is complicated.Based on the theories of support vector machines and wavelet support vector machines, two linear models for predicting concrete strength were proposed.Then two approaches are applied to predict the concrete strength.Furthermore, the predicted results are compared with actual measured.Results shows two methods were consist with measured ones. It is found that wavelet support vector machines can predict the strength of high strength concrete more accurately than the approach of support vector machines does.Proposed methods are quantitative and convenient analyzing approach with high accuracy.They are feasible in predicting the strength of high strength concrete.

关 键 词：混凝土 非线性方法 强度预测 

分 类 号：TU528.01[建筑科学—建筑技术科学]