作 者: (周亚勤); (杨建国); (刘凯强); (尤祥);
机构地区: 东华大学机械工程学院,上海201620
出 处: 《东华大学学报(自然科学版)》 2017年第4期515-519,540,共6页
摘 要: 在加工过程中,刀具磨损状况对零件的加工质量具有重要影响,精确预测刀具寿命是智能制造系统必须具有的关键功能之一.在分析数控铣刀寿命影响因素的基础上,引入极限学习机(ELM)算法模型,建立数控铣刀寿命预测模型.在寿命预测过程中,采用遗传算法(GA)对ELM模型的输入权值和隐含层阈值进行优化,建立基于GA-ELM的数控铣刀寿命预测模型,同时将其与基本BP神经网络、优化BP神经网络和基于粒子群改进的BP神经网络的预测结果进行对比分析.结果表明,基于GA-ELM的刀具寿命预测模型相比较于其他3种算法更加优越,是一种行之有效且精度高的刀具寿命预测算法. In the process of machining, the wear condition of the tool has an important effect on the machining quality of the parts. Accurately predicting the tool life is one of the key functions that the intelligent manufacturing system must'haVe. Based on the analysis of the influence factors of the life of CNC (computer numerical control) milling cutter, the extreme learning machine (ELM) algorithm model is introduced to establish the life prediction model of CNC milling cutter. In the process of life prediction, the input weight and hidden layer threshold of ELM model are optimized by genetic algorithm (GA), and the life prediction model of CNC milling cutter based on GA-ELM is established. The prediction results are compared and analyzed with basic BP neural network, BP neural network, based on particle swarm optimization BP neural network, the results show that the tool life prediction model based on GA-ELM is superior to the other three algorithms, which is an effective and accurate tool life prediction algorithm.