摘要
在分析了传统BP神经网络缺点的基础上,对其进行了改进。学习因子按照所设计的函数进行动态调整,增加了势态项,并在激活函数中加入了陡峭因子。将改进的BP神经网络与传统BP神经网络进行了对比,验证了改进BP神经网络的有效性,适合数控机床的故障诊断。
On the base of analyzing the traditional BP neural network＇s disadvantages,improvement method is given.Learning factor is dynamically adjusted according to the designed function,momentum term is added,sharpness factor is added in activation function.The improved BP neural network is compared with traditional BP neural network,the improved BP neural network＇s validity is verified and is suitable for NC machine＇s fault diagnosis.
出处
《制造技术与机床》
CSCD
北大核心
2011年第2期70-72,共3页
Manufacturing Technology & Machine Tool
基金
国家科技重大专项(2009ZX04014-014)
