BP神经网络预测环境对丝织物力学性能的影响  

Prediction for effect of environmental factors on silk fabric mechanical properties by BP neural network

作　　者：杨梦星[1] 张毅[1] 郑秋婷[1] 

机构地区：[1]天津工业大学纺织学部,天津300387

出　　处：《纺织学报》2013年第8期39-41,47,共4页Journal of Textile Research

摘　　要：为研究环境中湿度、温度和pH值以及它们之间的交互作用导致丝织物老化的问题,建立BP神经网络模型,预测环境因素对丝织物力学性能的影响。首先采用L9(34)正交试验设计湿度、温度和pH值的交互环境,并将丝织物置于交互环境中进行24 h的连续老化处理。再使用3365型Instron万能材料试验机测试丝织物老化后的断裂强力,并将断裂强力值作为BP神经网络的训练样本,建立3层的神经网络模型进行迭代训练以及预测。结果显示训练输出值与实际试验值的相对误差均小于0.005%,表明BP神经网络模型可有效地对丝织物的力学性能进行预测。For the problem that humidity,temperature,PH value and their interaction can accelerate the ageing of silk fabrics,a BP neural network model was developed to predict the effect of environmental factors on the mechanical properties of silk fabrics.Firstly,L9（34） orthogonal experiment was employed to design the interactive environment of humidity,temperature and PH value,and silk fabric were placed in the interactive environment and subjected to ageing treatment for 24 h.Secondly,the type 3365 Instron universal material testing machine was used to test the breaking strength of the silk fabric after ageing treatment,and the breaking strength value obtained was used as the BP neural network training sample,and then a three-layer neural network was established to perform iterative training and prediction.The result showed that the relative errors of the training output values and the actual experimental values are smaller than 0.005%,indicating that BP neural network can predict the mechanical properties of silk fabrics effectively.

关 键 词：BP神经网络 丝织物 环境因素 力学性能 

分 类 号：TS141.9[轻工技术与工程—纺织材料与纺织品设计]