| 基于非线性优化理论的混凝土材料配合比研究 |
| |
| 引用本文： | 刘富玲.基于非线性优化理论的混凝土材料配合比研究[J].混凝土,2012(4):72-73,76. |
| |
| 作者姓名： | 刘富玲 |
| |
| 作者单位： | 河南城建学院土木工程系建筑材料教研室 |
| |
| 摘 要： | 建立了高强混凝土的强度预测的非线性优化模型。并将该模型计算结果与实测混凝土28 d抗压强度进行比较。改用十进制遗传算法在训练过程中搜索最优超参数,形成遗传-组合核函数高斯过程回归算法,并编制了相应的计算程序,研究结果表明:与单一核函数高斯过程回归算法和支持向量回归(SVR)算法相比,提出的遗传-组合核函数高斯过程回归算法显著提高了预测精度,预测结果与实测结果吻合较好,具有较高的预测精度,为高强混凝土的强度预测提供了一条新途径。
|
| 关 键 词： | 混凝土 强度预测 组合核函数 遗传算法 |
Strength prediction of high strength concrete using forecast optimized nonlinear model |
| |
| LIU Fu-ling.Strength prediction of high strength concrete using forecast optimized nonlinear model[J].Concrete,2012(4):72-73,76. |
| |
| Authors: | LIU Fu-ling |
| |
| Affiliation: | LIU Fu-ling(Civil Engineering Department of Henan University of Urban Construction Building Materials Department,Pingdingshan 467044,China) |
| |
| Abstract: | To make the strength forecast of high strength concrete under influence of several factors exact,the model of optimized method of Genetic algorithm and its learning algorithms are recommended.Then the model is applied to predict the strength of high strength concrete.Furthermore,we contrast it to results of actual measured.Genetic algorithm coded in decimal system was used to optimize the hyper parameters of the GPR then formed the GA-CKGPR algorithm,and the corresponding code was programmed.The analysis results show that the GA-CKGPR algorithm obviously improved the prediction precision than SVR and standard GPR respectively,so it is feasible in predicting the strength of high strength concrete. |
| |
| Keywords: | concrete strength forecast genetic algorithm time series analysis |
| 本文献已被 CNKI 等数据库收录！ |
|
