摘要
特征识别技术是CAD/CAPP/CAM集成的关键技术之一。针对现有特征识别方法的缺点,提出了一种基于属性邻接图和神经网络的加工特征识别方法。该方法采用属性邻接图的方法对产品数据交换标准(standard for the exchange of product model data,STEP)文件进行预处理,生成神经网络的输入矢量,同时用遗传算法对BP神经网络进行了优化;最后采用了一个实例对该方法进行了说明。通过仿真验证了该方法在特征识别中的有效性,为特征识别提供了一种新方法。
Machining feature recognition technology is one of the key technologies of the integration of CAD /CAPP / CAM. According to the shortcomings of existing feature recognition method,this paper presents a method of feature recognition based on attribute adjacency graph and neural network. Firstly,it uses method of the attribute adjacency graph to preprocess the STEP file and then get the input vector of neural network. Secondly,it uses genetic algorithm to optimize the BP neural network. Finally,it takes an example to illustrate the method. The results of simulation show the effectiveness of this method in feature recognition,it provides a new way for feature recognition.
出处
《机电一体化》
2015年第10期7-11,37,共6页
Mechatronics
关键词
特征识别
属性邻接图
神经网络
STEP文件
feature recognition
attribute adjacency graph
neural network
STEP file
