PE–HD/PE–LD振动挤出预测模型及其应用  

Vibration Extrusion PE–HD/PE–LD Forecasting Model and Its Application

作　　者：陈学锋[1] 徐言生[1] 胡建国[1] 许中明[1] 殷小春[2] 

机构地区：[1]顺德职业技术学院机电工程学院,广东顺德528300 [2]华南理工大学聚合物新型成型装备国家工程研究中心,聚合物成型加工工程教育部重点实验室,广州510641

出　　处：《工程塑料应用》2015年第2期55-59,共5页Engineering Plastics Application

基　　金：广东省自然科学基金资助项目(S2012010010199);(华南理工大学)聚合物成型加工工程教育部重点实验室开放课题基金资助项目(2012001);顺德区科技计划项目(20120202087);顺德区容桂科技计划项目(RGKJJH2013008)

摘　　要：基于BP人工神经网络,研究高密度聚乙烯(PE–HD)、低密度聚乙烯(PE–LD)材料的振动挤出加工过程,建立螺杆转速、振动及口模诸因子与挤出胀大、压力和功率等输出特性参数间精确、高效、简洁的非线性映射关系,为挤出加工参数的优化和挤出加工质量的控制提供依据。结果证明,基于BP神经网络模型能够很好地预测聚合物挤出加工特性参数,所建网络具有精确、高速、自适应等特点。Based on BP neural network,the vibration extrusion process of high density polyethylene（PE–HD） and low density polyethylene（PE–LD） were studied. A high efficiency,simple and precise nonlinear mapping relationship were established,which were about the screw speed,vibration and die as the input factor and the extrusion swell,pressure and power as the output factor relations. Provide the basis for the optimization of extrusion process parameters and controlling the quality of extrusion processing. The experimental results show that the BP neural network model can well predict the processing properties of polymer extrusion parameters. The neural network model has the advantages of high precision,high speed,adaptive characteristics.

关 键 词：人工神经网络 BP算法 振动 挤出特性 

分 类 号：TQ320.52[化学工程—合成树脂塑料工业]