神经元网络技术方法研究PET高速纺丝过程中的细颈点

STUDY ON NECKING POINT IN PET HIGH SPEED SPINNING WITH ARTIFICIAL NEURAL NETWORKS

首次将神经元网络技术运用于ＰＥＴ高速纺丝细颈点的研究中,同时比较了神经元网络预测结果与文献报导数据。并对神经元网络方法和数学模拟方法进行讨论,分析了两种方法的优缺点,分“互补法”的建立奠定了基础。

Artificial neural networks(ANNs)were used to study on necking point in PET high speed spinning for the first time. The ptedicted results with ANNs were compared with data reported by literature. The advantages and disadvantages were analyzed for ANNs and mathematical simulation. A foundation was kid on the establishment of cross over method.