【作者】 陈晓华；

【导师】 李江平；

【作者基本信息】 广东工业大学 ， 高分子化学与物理， 2012， 硕士

【摘要】 随着国民经济的高速发展,也带动了塑料工业的飞速发展,塑料制品使用范围在不断扩大。但是在注塑成型过程中出现的各种问题特别是成型薄壁塑料制品时产生的翘曲变形和收缩率过大等,如何减少、防止制品成形中时常发生影响产品质量的问题,是经常困扰工程师的一大难题。在手机行业中,绝大部分的部件都是由塑料制品组成,塑料成型CAE技术是当前塑料加工行业研究的热点,但是目前注塑成型CAE技术依然存在着一系列的问题,特别是在成型手机外壳这种薄壁塑料制品的时候。本文工作主要包括以下几个方面的内容：1.本文简要概述了薄壁注塑成型技术,介绍了薄壁件翘曲变形产生的相关机理,以及聚合物流变学的基础知识,进而分析影响翘曲变形量和体积收缩率的因素,并指出了减小薄壁件翘曲变形和收缩率的多种措施,由此讨论并且确定了薄壁件翘曲变形的研究方法。2.注塑制品翘曲变形量和体积收缩率受工艺参数的的影响进行综合评价。并且指出翘曲量和体积收缩受工艺参数控制的影响不可忽视,最后得到合适的工艺参数组合使翘曲变形分析和体积收缩率最小。3.基于神经网络建立了从注塑工艺参数到翘曲变形量和体积收缩率的非线性映射关系。利用正交试验获得的数据作为人工神经网络的训练样本,得到输入为工艺参数、输出为翘曲变形量和收缩率的人工神经网络模型,并且通过对样本的检验,检验了ANN(Artifidal Neural Network)模型的准确性,为参数优化及翘曲变形和收缩率的预测做好准备。4.基于人工神经网络和正交试验的工艺参数优化。我们在工艺参数取值范围内采用ANN模型代替CAE软件进行数值模拟试验,并且结合正交试验法对工艺参数做优化使得翘曲变形量和收缩率更小。论文工作表明：将正交试验和神经网络以及数值模拟三者结合,用于注塑过程参数优化可非常明显的减少优化工艺参数的时间并且提高工艺设定的效率,同时在数值模拟试验次数一定的条件下能够获得比单一使用正交试验和数值模拟方法更为精确的结果。本论文对薄壁注塑翘曲变形作了多工艺参数影响的综合分析减少了单个因素分析的片面性,有利于对薄壁注塑件翘曲变形和收缩率问题的深入探讨,在保证分析精度一定的前提下,明显节省了工艺制定的时间,提高了工艺设计的工作效率,缩短生产时间,提高制件质量。更多还原

【Abstract】 With the rapid development of the national economy has also led to the rapid development of the plastics industry and the range use of plastic products in expanding. But various problems in the injection molding process, especially when molding thin-walled plastic products, warpage and shrinkage is too large, engineers often plagued by a major problem that how to reduce, to prevent the products shaping often affect product. In the mobile phone industry, the vast majority of parts are composed of plastic products, current research focus of the plastics processing industry is plastic molding CAE technology, but there are still a range of issues in the injection molding CAE technology, especially in the forming cell phone case thin-walled plastic products.This work includes the following aspects of the content:1. This paper provides an overview of the thin-wall injection molding technology and warpage related mechanism of the thin-walled parts, as well as the basics of polymer rheology, and then analyze the factors that affect the amount of warpage and volume shrinkage, pointed out variety of measures to reduce small thin-walled parts warpage and shrinkage, thus discussed and determined some research methods of the thin-walled warpage.2. Evaluation of multiple process parameters influence on injection molding warping deformation quantity and size of shrinkage ratio. With cell phone casing parts as the research object and building simulation model, arranged experiment by taguchi test method, obtain the warpage amount and volume contraction rate of the test data by simulate and analysis the injection molding of plastic parts. Different process parameters play an essential role on the warpage and shrinkage control, at the same time we can get the minimum combinations of process parameters of the warpage.3. Based on neural network establish a nonlinear mapping relationship from the injection molding process parameters to the warpage amount and volume shrinkage, taguchi experiment dated as an artificial neural network training samples. the input is process parameters, the out put is warpage and shrinkage of the artificial neural network model, testing ANN (Artifidal, Neural Network) the accuracy of the model by inspection of samples, and be prepare for the parameter optimization and warpage and shrinkage forecast.4. Based on artificial neural network and the process parameters optimization of the taguchi experiment. we are using ANN model instead of the CAE software with the combination of the taguchi experiment method to simulate the test within the range of process parameters, get smaller warpage amount of process parameters by futher optimization. The thesis work has shown that:we can significantly shorten the time to optimize the process parameters and to increase the efficiency of the process design by combined with the taguchi experiment, the neural network and numerical simulation of the three combination for the injection molding process parameters optimization, we can obtain more accurate results than the single use of orthogonal experiments and numerical simulation methods under certain conditions of the number of numerical simulation test.This paper made several process parameters of the combined effects of analysis to the shell injection molding warpage, to avoid the one-sidedness of a separate analysis. Use taguchi experiment, the neural network and the numerical simulation method for the optimization of mobile phone shell injection molding process parameters, under certain preconditions to ensure the accuracy of analysis, improve the efficiency of the process design work, shorten production time and improve part quality.更多还原