摘要
磨料水射流是一种先进的绿色加工工具。为了获得较高的表面精度,必须精确控制水射流加工的各种工艺参数。一种可行的办法是建立水射流加工过程中的主要参数之间的非线性关系,通过加工速度的补偿来间接控制工件表面质量。本文基于人工神经网络理论,建立了水射流加工过程的神经网络模型。在获取大量样本数据的基础上,对网络模型进行训练。经过训练的水射流加工神经网络模型可以准确预测与给定压力、磨料流量、工件厚度、期望的表面粗糙度所对应的切割速度,实验验证获得满意结果。
Abrasive Water Jet is one of the advanced green machining tools. In order to obtain high surface quality of the product, the parameters of abrasive water jet machining process must be precisely controlled. A practical technique is to establish the non-linear relationship among the main parameters of the abrasive water jet cutting process, thus the surface quality of the part can be indirectly controlled by adjusting the cutting speed. Based on the artificial neural network (ANN), an ANN model for the abrasive water jet cutting process was proposed.This model was trained using sample data set. The trained model can precisely predict the cutting speed at the given pressure, flow rate of abrasive, thickness of the material and roughness of the part to be made.
出处
《四川大学学报(工程科学版)》
EI
CAS
CSCD
北大核心
2005年第6期155-159,共5页
Journal of Sichuan University (Engineering Science Edition)
基金
四川省人事厅学术带头人培养资金(0422006)
四川省科技厅资助项目(05202073)
关键词
磨料射流
模型
人工神经网络
abrasive water jet
modeling
artificial neural network