检索规则说明:AND代表“并且”;OR代表“或者”;NOT代表“不包含”;(注意必须大写,运算符两边需空一格)
检 索 范 例 :范例一: (K=图书馆学 OR K=情报学) AND A=范并思 范例二:J=计算机应用与软件 AND (U=C++ OR U=Basic) NOT M=Visual
机构地区:[1]第二炮兵工程学院,陕西西安710025 [2]西安电子科技大学技术物理学院,陕西西安710071 [3]国防科技大学计算机学院,湖南长沙430071
出 处:《稀有金属材料与工程》2007年第A02期283-285,共3页Rare Metal Materials and Engineering
基 金:陕西省自然科学基金(2004E1-13)
摘 要:采用共混法制备镁铝水滑石/聚合物复合材料。选用BP神经网络建模,输入输出神经元分别由制备时工艺参数和预测指标来确定。为了减小振荡、加快网络收敛速度,修改权值的附加项引用模拟退火法、根据代价函数调整学习速率梯度、采用隐含层结点之间相关性来合并结点、用分散度消除多余结点等措施,获得适当大的隐含层结点数。仿真结果表明:使用改进的BP神经网络模型,可以有效地进行此复合材料的性能预测。Mg,Al-hydrotalcite/polyethylene nanocomposite samples were synthesized by the co-mixing method at atmospheric pressure. The prediction model for the stretch strength of the artificial synthetic Mg,Al-hydrotalcite/polyethylene nanocomposite under varied process parameters based on improved artificial back-propagation (BP) neural networks was developed. And the non-linear relationship between the nanocomposite stretch strength and the technology factors, such as surface active agent (sodium oleate) adding amount, Mg,Al-hydrotalcite adding amount, pH value of the reaction medium, was established. Moreover, in order to accelerate the converging rate, avoid the local minimum and reduce the oscillation of BP neural networks, simulated annealing was used to modify the addition items of weight values; and the gradient descent of the learning rate was adjusted according to the cost function; besides, a proper hidden-layer node number was obtained by the methods that the nodes were merged based on the interrelated relationship of the hidden-layer nodes and redundant nodes were eliminated by using dispersity. Finally, the results show that the improved back propagation neural networks model is very efficient for the stretch strength predication of Mg,Al-hydrotalcite/polyethylene nanocomposite.
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在链接到云南高校图书馆文献保障联盟下载...
云南高校图书馆联盟文献共享服务平台 版权所有©
您的IP:216.73.216.63