基于神经网络技术的注射螺杆熔融性能  

Melting Performance of Injection Screw based on Natural Network

作　　者：董英[1] 金志明[1] 

机构地区：[1]北京化工大学机电工程学院,北京100029

出　　处：《塑料》2013年第4期24-27,共4页Plastics

摘　　要：熔融性能是指注射螺杆在塑化过程中对物料的熔融能力,可以用熔融曲线来表示。文章利用自行开发的可视化实验装置,测量得到了不同加工工艺条件下的固相分布,以此为基础建立了基于BP神经网络的固体床宽度分布函数的预测,并利用实验结果对所建网络模型的预测能力进行验证,结果表明该网络模型的预测效果好。利用所建立的模型研究分析了螺杆转速、背压、注射行程等工艺条件对固相宽度分布函数的影响。Melting performance referred to melting capacity of injection screw in plasticizing process and expresses in melting curve. The widths of solid bed distribution function with different process conditions were measured by means of visual experimental system. A back propagation natural network was trained to predict the width of solid bed distribution function,and experimental results were used to check the performance of the neural network. The results showed that the neural network had a good simulation. The influences of the width of solid bed with different process conditions were analysed. According to the model established, the effect of different process conditions such as screw rotation speed, back pressure and injections stroke on the width of solid bed waw analysed.

关 键 词：注射成型 注射螺杆 BP神经网络 熔融性能 固相宽度分布函数 

分 类 号：TQ320.66[化学工程—合成树脂塑料工业]