摘要
针对核电数控轮槽铣床可靠性数据样本量复杂,传统可靠性分析方法无法进行可靠性分析或分析误差较大等问题,采用神经网络对核电数控轮槽铣床进行可靠性分析,通过建立神经网络可靠性模型,可以准确的确定各子系统可靠度及对应的平均无故障时间与系统可靠度之间的关系,将训练结果与实际结果相比最大误差仅为0.089,可见神经网络大幅度提高了机床的可靠度分析精度。为系统可靠性分析和系统维修决策的研究提出了新的思维路径,具有很好的应用前景。
The reliability data of nuclear power CNC wheel groove milling machine is so few that the reliability can not be predicted by traditional reliability prediction method; or the reliability prediction error is large. A reliability prediction method of nuclear power CNC wheel groove milling machine based on the neural network is proposed here. Using the neural network to nuclear power CNC wheel groove milling machine reliability analysis, through the establishment of neural network reliability model, which can accurately determine the reliability of each subsystem and the corresponding MTBF and system reliability, the relationship between the training results and the actual results will be compared to the maximum error which is only 0.089, and visible neural network greatly improves the reliability analysis of the machine tool accuracy. For system reliability analysis and system maintenance decision research it puts forward a new thinking path, have a good applied foreground.
出处
《机械设计与制造》
北大核心
2013年第9期134-136,139,共4页
Machinery Design & Manufacture
基金
国家重大科技专项课题(2011ZX04002-081)
关键词
核电数控轮槽铣床
神经网络
平均无故障时间
系统可靠度
可靠性分析
Nuclear Power NC Wheel Groove Milling Machine
Neural Network
Mean Time Between Failures
System Reliability
Reliability Analysis
