振动诱导聚合物塑化过程中熔体温度分布的预测

在振动力场诱导聚合物塑化成型作用下,建立了聚合物熔融挤出过程中的熔体温度分布模型,研究了模头温度、振动力场的振幅、频率等工艺参数对挤出过程中熔体温度的影响。提出了基于多项式和高斯RBF核函数变换的两种非线性岭回归模型(PT-RR和GRBF-RR),并对具有非线性、非等温、强耦合特征的熔融过程熔体温度分布进行研究。该建模方法实现了多变量输入样本的高维特征空间非线性映射与重构,充分挖掘了多影响因素之间的耦合信息。仿真实验结果表明了PT-RR和GRBF-RR模型的有效性,其回归预测值与实验测量值之间的相关系数均值分别为0.9940和1。由于GRBF-RR模型取得了满意的模型精度,本文基于对GRBF-RR模型的数值模拟分析了各影响因素对熔体温度分布的交叉耦合影响,表征了聚合物熔融挤出成型过程中熔体温度分布的规律。该项研究可为提高精密挤出制品质量及优化配置各工艺参数提供决策依据。

Prediction of melt temperature distribution in vibration induced polymer plasticization

Melt temperature distribution models for melting and plasticizing process of polymer materials were presented in vibration field induced extrusion, and the influences of process parameters such as die temperature, vibration amplitude and frequency on the melt temperature were investigated. Two kinds of ridge regression models based on the nonlinear transforms of polynomial and Gaussian RBF (PT-RR and GRBF-RR) were established to predict the melt temperature distribution with the nonlinear, non-isothermal properties and strong-coupling extrusion. The two models have fulfilled the nonlinear mapping and reconstruction of high-dimension feature space from multi-variable input samples, and obtained the coupling relations among multi-factor influences by the numerical simulation based on GRBF-RR. The simulation and experimental results show that the two models are valid and the correlation coefficients between the predicted values and that measured values are 0.994 0 and 1 for PT-RR and GRBF-RR, respectively. As the GRBF-RR can offer higher model precision, it was used to illuminate the influences of process parameters on the melt temperature distribution. Obtained results demonstrate that the model can provide a decision support for the quality control and process parameter optimization.