| 基于IFOA-RBF算法的混凝土抗压强度预测 |
| |
| 引用本文： | 徐富强,陶有田.基于IFOA-RBF算法的混凝土抗压强度预测[J].巢湖学院学报,2014(6):7-11. |
| |
| 作者姓名： | 徐富强 陶有田 |
| |
| 作者单位： | 1. 巢湖学院数学系,安徽 巢湖,238000
2. 巢湖学院数学系，安徽 巢湖 238000; 中国科学技术大学数学科学学院博士后流动站，安徽 合肥 230026; 安徽富煌钢构有限公司博士后工作站，安徽 巢湖 238076 |
| |
| 基金项目： | 安徽高校省级自然科学研究项目，大学数学基础课程教学团队 |
| |
| 摘 要： | 为了提高混凝土抗压强度预测精度,利用改进果蝇优化算法(IFOA)优化RBF神经网络的参数Spread值,建立IFOA-RBF预测模型用于混凝土抗压强度预测。模型以UCI数据库中的Concrete Compressive Strength数据集为例,以每立方混凝土中的水泥、高炉矿渣粉、粉煤灰、水、减水剂、粗集料和细集料的含量以及置放天数为网络输入,混凝土抗压强度值作为网络输出,进行仿真测试,并将结果与参考文献中的其它方法比较。结果表明:优化后的RBF网络既体现了广泛映射能力,又明显地提高了网络的泛化能力。验证了IFOA-RBF模型在混凝土抗压强度预测中的有效性。
|
| 关 键 词： | 果蝇优化算法 RBF神经网络 参数优化 混凝土抗压强度 |
ON THE PREDICTION OF CONCRETE COMPRESSIVE STRENGTH BASED ON THE ALGORITHM OF IFOA-RBF |
| |
| XU Fu-qiang,TAO You-tian.ON THE PREDICTION OF CONCRETE COMPRESSIVE STRENGTH BASED ON THE ALGORITHM OF IFOA-RBF[J].Chaohu College Journal,2014(6):7-11. |
| |
| Authors: | XU Fu-qiang TAO You-tian |
| |
| Institution: | XU Fu-qiang, TAO You-tian (1 Department of Mathematics, Chaohu College, Chaohu Anhui 238000; 2 Postdoctoral Research Station for School of Mathematics Sciences, USTC, Hefei Anhui 230026; 3 Postdoctoral Workstation for Anhui Fuhuang Steel Structure Co. Ltd., Chaohu Anhui 238076) |
| |
| Abstract: | In order to improve the accuracy of predicting the concrete compressive strength,an IFOA-RBF forecasting model for predicting the concrete compressive strength is established through improving the IFOA and optimizing the Spread value of RBF neural network. The model uses the concrete compressive strength data in UCI database as an example, and the simulation testing is carried out by setting the content of cement, blast furnace slag, fly ash, water, water reducer, coarse aggregate and fine aggregate in per cubic concrete and using days of their placement as the network input, and meanwhile the concrete compressive strength value is used as the network output.Then this paper compares the results with the conclusion of the references. The results show that： the optimized RBF network not only embodies the extensive mapping ability, but also significantly improves the network generalization ability. The validity of the IFOA-RBF model in concrete compressive strength prediction is verified. |
| |
| Keywords: | FOA RBF neural network parameter optimization concrete compressive strength |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
