摘要
在分析纺纱原理的基础上,采用人工神经网络建立7 个不同的模型,预测精梳毛纺的纱线质量和纺纱性能,分别是纱线不匀、粗细节、断裂强力、强力不匀、断裂伸长和断头率。其中断头率由于其复杂的成因,采用了组合神经网络建模。采用工厂实际生产数据进行验证,前6个指标的预测值与实测值之间的相关系数的平方均超过0.9,断头的预测效果相比而言比较差,但相关系数的平方也超过了0.8,表明人工神经网络技术在精梳毛纺纱线预测方面有很大的应用前景。
Based on the analysis of spinning principle, we built seven different neural networks to predict yarn quality and spinning performance, i.e. yarn evenness, thin places, thick places, yarn tenacity and its variation, elongation at break and the rate of ends-down. Due to complex cause of ends-down, compound neural network was developed to predict the rate of ends-down. After validated by using the data from a mill, the neural network gave good results. The square of correlate coefficient (R 2) between predicted value and measured value of the seven indices except the rate of ends-down exceed 0.9. The R 2 of the rate of ends-down is 0.86, giving good prediction. It shows the neural network provides a powerful tool for worsted yarn prediction.
出处
《东华大学学报（自然科学版）》
CAS
CSCD
北大核心
2005年第1期66-71,共6页
Journal of Donghua University(Natural Science)
关键词
神经网络
纱线预测
多层感知器
neural network, worsted yarn qualities prediction, multi-layer perceptron
