摘要
BP神经网络在旋转机械故障诊断方面取得了巨大成果,但因其自身的局限性,还存在网络结构难以确定、学习时间较长、易收敛于局部最优解等问题。采用遗传算法对BP神经网络进行优化,并将其应用于转子故障诊断,能有效克服以上缺点,效果较好。
BP neural network fault diagnosis of rotating machinery made in terms of a great achieve- ment, but because of its own limitations, there are still difficult to determine the structure of the network, such as that it will take a long time to learn, and it is easy to converge to the optimal solution and so on. The paper optimized rotor fault diagnosis by taking genetic BP neural network algorithm and applied it to rotor fault diagnosis, which can effectively overcome the above drawbacks and achieve good results.
出处
《重庆理工大学学报（自然科学）》
CAS
2015年第7期44-48,共5页
Journal of Chongqing University of Technology：Natural Science
基金
国防项目"群车加油车效能提升智能控制技术研究"(YX214J038)
关键词
BP神经网络
遗传算法
转子故障
BP neural networks
genetic algorithms
rotor fault
