|
|||||
|
|
| 基于BP神经网络的充填料浆流变参数预测分析 | |
| 引用本文： | 邓代强,朱永建,李健,张友轩.基于BP神经网络的充填料浆流变参数预测分析[J].武汉理工大学学报,2012,34(7):82-87. |
| 作者姓名： | 邓代强 朱永建 李健 张友轩 |
| 作者单位： | 1. 长沙矿山研究院有限责任公司采矿工程中心,长沙410012;湖南科技大学煤矿安全开采技术湖南省重点实验室,湘潭411201;国家金属采矿工程技术研究中心,长沙410012 2. 湖南科技大学煤矿安全开采技术湖南省重点实验室,湘潭,411201 3. 北京科技大学土木与环境工程学院,北京,100083 4. 长沙矿山研究院有限责任公司采矿工程中心,长沙,410012 |
| 基金项目： | 国家“十一五”科技支撑计划(2008BAB32B01);湖南省自然科学基金(11JJ5030);煤矿安全开采技术湖南省重点实验室开放基金(201108) |
| 摘 要： | 为了合理分析充填料浆在长距离管道中的输送阻力,基于流变参数对输配管网设计的重要性,在全面评估不同因素对流变性能影响的程度后,采用BP神经网络原理,建立起干料中水泥含量X1、料浆浓度X2、料浆坍落度X3、料浆容重X4对流变参数(屈服应力Y1、粘性系数Y2)影响的函数模型.此BP网络为4-Hn-2结构,隐层和输出层分别用tansig、purelin函数传递,利用Levenberg-Marquardt优化算法trainlm训练网格.计算结果表明:该模型在预测充填料浆屈服应力Y1和粘性系数Y2中适应性较强,误差也在可控范围之内,可为充填管网布设及输配系统沿程阻力分析提供可靠依据. |
| 关 键 词： | 充填料浆 管道输送 流变性能 BP神经网络 预测模型 |
Rheology Parameter Forecast Analysis of Filling Slurry Based on BP Neural Network |
|
| DENG Dai-qiang , ZHU Yong-jian , LI Jian , ZHANG You-xuan.Rheology Parameter Forecast Analysis of Filling Slurry Based on BP Neural Network[J].Journal of Wuhan University of Technology,2012,34(7):82-87. | |
| Authors: | DENG Dai-qiang ZHU Yong-jian LI Jian ZHANG You-xuan |
| Affiliation: | 1(1.Center of Mining Engineering,Changsha Institute of Mining Research Co,Ltd,Changsha 410012,China; 2.Hunan Provincial Key Laboratory of Safe Mining Techniques of Coal Mines,Hunan University of Science and Technology,Xiangtan 411201,China;3.National Engineering Research Center for Metal Mining,Changsha 410012,China; 4.School of Civil and Environmental Engineering,University of Science and Technology Beijing,Beijing 100083,China) |
| Abstract: | In order to analyze the transportation resistance of filling slurry in long distance pipeline reasonably,based on the importance of rheological parameters for transmission and distribution pipe network design,influence degrees of rheological property caused by different factors were estimated in all directions.By using BP neural network theory,the influence function model of rheological parameters(slurry yield stress Y1,viscous coefficient Y2) caused by cement content X1,slurry concentration X2,slurry slump extent X3 and slurry density X4 in dry material was established,which is a 4-Hn-2 network structure,in which hidden layer and output layer was transferred by Tansig and purelin function respectively.Levenberg-Marquardt optimization algorithm,trainlm,was also used to train network.The results show that the model can effectively predict the filling of the yield stress of the slurry Y1 and viscous coefficient Y2,and the error is also controlled within the scope.So we can provide a theoretical basis for the filling pipe network layout and transmission and distribution network analysis of resistance along the way. |
| Keywords: | filling slurry pipeline transportation rhyeology performance BP neural network forecast model |
| 本文献已被 CNKI 万方数据 等数据库收录！ | |