基于ANN代理模型的单螺杆计量段结构参数优化  

Optimization of structural parameters of single-screw metering section based on ANN surrogate model

作　　者：王超元 陈欣[1] 林增 祁纪浩 庞志威 沙金[1] WANG Chaoyuan;CHEN Xin;LIN Zeng;QI Jihao;PANG Zhiwei;SHA Jin(School of Mechanical and Power Engineering,East China University of Science and Technology,Shanghai 200237,China)

机构地区：[1]华东理工大学机械与动力工程学院,上海200237

出　　处：《中国塑料》2025年第3期95-101,共7页China Plastics

摘　　要：在挤出机单螺杆计量段二维解析建模的基础上,采用交叉验证方法构建人工神经网络(artificial neural network,ANN)模型并对其进行了超参数优化,以有效地映射挤出机工作条件和结构参数与生产率和功耗之间的复杂非线性关系。提出利用ANN代理模型,结合NSGA-Ⅱ(non-dominated sorting genetic algorithmⅡ)算法对螺杆计量段的结构参数进行多目标优化,并通过TOPSIS(technique for order preference by similarity to an ideal solution)法得到最优生产率和功耗组合的结构参数。相关工作对单螺杆计量段结构参数的智能化设计具有理论指导意义。Based on the two-dimensional analytical modeling for the single-screw metering section of an extruder,a crossvalidation method was employed to construct an artificial neural network(ANN) model and optimize its hyperparameters to effectively map the complex nonlinear relationship between the working conditions and structural parameters of the extruder and the productivity and power consumption.A multi-objective optimization of the structural parameters of the screw metering section was proposed using the ANN surrogate model combining with the NSGA-Ⅱ (non-dominated sorting genetic algorithm Ⅱ) algorithm,and the structural parameters of the optimal combination of productivity and power consumption were obtained through the TOPSIS(technique for order preference by similarity to an ideal solution) method.The relevant work is of theoretical guidance significance for the intelligent design of the structural parameters of the single screw metering section.

关 键 词：单螺杆结构参数 人工神经网络 多目标优化 NSGA-II 

分 类 号：TQ320.663[化学工程—合成树脂塑料工业]