利用BP神经网络法，建立了再生混凝土的抗压强度预测模型。该模型采用了3层网络结构模式，输入层采用再生混凝土的配合比数据，输出层为再生混凝土，置放7、28、56、90 d的强度数据，模型的转移函数均采用单极性Sigmoid函数。由于Sigmoid函数值为[0,1]，因此，对再生混凝土输入数据进行归一化处理。最后，设计了33块再生混凝土的抗压强度试验，利用试验数据对网络模型进行测试，测试的结果证实了该模型对再生混凝土的强度预测值与实际测试结果基本相符。
Abstract
The forecast model about the compressive strength of recycled aggregate concrete was set up using the method of BP neutral network. The model was threelayer network structure pattern. The inputtinglayer was the mix ratio of recycled aggregate concrete. The outputtinglayer was the strengths data of the recycled concentrate in 7, 28, 56, 90 d. The model’s transfer function was Sigmoid function.Because the numerical value of Sigmoid was in [0,1]. This article turned the inputting data into the data in [0,1]. At last, the experimental work for the compressive strength of recycled aggregate concrete was designed and performed with 33 pieces. The model was trained and tested by experimental data.It showed that forecast result was match with test results and it could be referenced for experimentation and manufacture.
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1]肖建庄,李佳彬,兰阳. 再生混凝土技术最新研究进展与评述[J]. 混凝土,2003(10):17-20. [2]邢振贤,周日农. 再生混凝土的基本性能研究[J]. 华北水利水电学院学报,1998,19(2):30-32. [3]柯国军,张育霖,贺涛,等. 再生混凝土的实用性研究[J]. 混凝土,2002(4):47-48. [4]孔德玉. 天然与再生骨料混凝土水灰比统一定则(1)—粗骨料强度的影响[J]. 建筑材料学报,2003,6(2):129-134. [5]李占印. 再生粗骨料混凝土的性能[D]. 西安:西安建筑科技大学,2003. [6]邓旭华. 水灰比对再生混凝土强度影响的试验研究[J]. 混凝土,2005,28(2):46-48. [7]Limbachiya M C, Leelawat T, Dhir R K. Use of recycled concrete aggregate in highstrength concrete[J]. Materials and Structures, 2000, 33(9): 574-580. [8]Tavakoli M, Soroushian P. Strengths of recycled aggregate concrete made using fielddemolished concrete as aggregate[J]. ACI Mater J, 1996, 93(2): 178-181. [9]陈德玉,吕淑珍,刘欢,等. 掺合料对再生混凝土强度和耐久性的影响[J]. 粉煤灰综合利用,2009(6):8-10. [10]李佳彬,肖建庄,黄健. 再生粗骨料取代率对混凝土抗压强度的影响[J]. 建筑材料学报，2006,9(3):297-301. [11]许岳周. 再生骨料混凝土级配对强度的影响研究与数值模拟[D]. 厦门:厦门大学,2007. [12]Yi Jianqiang, Wang Qian, Zhao Dongbin, et al. BP neural network predictionbased variableperiod sampling approach for networked control systems[J]. Applied Mathematics and Computation, 2007, 185( 2): 976-988. [13]Yu Shiwei, Zhu Kejun, Diao Fengqin. A dynamic all parameters adaptive BP neural networks model and its application on oil reservoir prediction[J]. Applied Mathematics and Computation, 2008, 195(1): 66-75. [14]高宁. 基于BP神经网络的农作物虫情预测预报及其MATLAB实现[D]. 合肥:安徽农业大学,2003. [15]Wu W, Feng G R, Li Z X, et al.Convergence of an online gradient method for BP neural networks[J]. IEEE Trans on Neural Neworks, 2005, 16: 533-540. [16]Li Z X, Wu W, Tian Y L. Convergence of an online gradient method for feedforward neural networks with stochastic inputs[J]. Journal of Computational and Applied Mathematics, 2004, 163(1): 165-176. [17]韩力群. 人工神经网络理论、设计及应用[M]. 北京:化学工业出版社,2007. [18]丛爽. 面向MATLAB工具箱的神经网络理论与应用[M]. 合肥:中国科学技术大学出版社,2003. [19]闻新,周露. MATLAB神经网络仿真与应用[M]. 北京:科学出版社,2003. [20]飞思科技产品研发中心. 神经网络理论与MATLAB7实现[M]. 北京:电子工业出版社,2005.
基金
收稿日期：2010-01-05修回日期：2010-06-10基金项目：衢州市科技局项目(20061048)作者简介：廖小辉(1978—)，讲师，博士生。*黄新(通信作者)，教授。Email: huangxin@njfu.com.cn。引文格式：廖小辉，黄新，施俊玲,等. 基于BP网络的再生混凝土抗压强度的预测模型[J]. 南京林业大学学报:自然科学版,2010,34(5):105-108.
PDF(607419 KB)
