摘要
为了提高齿轮减速器故障诊断结果的正确率,研究了一种基于神经网络优化FPN的齿轮减速器障诊断新算法.利用BP算法具有自适应学习的特点,在确定FPN相关网络参数原始数值的情况下,对FPN网络参数进行优化.利用BP算法在FPN网络基础上,对齿轮减速器故障样本进行学习训练,使FPN参数数值逐渐向真值靠近.实例结果表明:新算法对齿轮减速器中的单一或多种故障诊断非常有效,故障诊断结果准确率明显提高,说明优化算法的有效性与正确性.
In order to improve the accuracy of fault diagnosis result of gear reducer,this paper researches a fault diagnostic new method based on neural network optimized FPN.The BP algorithm has the characteristics of adaptive learning,and the original value of FPN related network parameters is optimized.Based on the FPN network,the BP algorithm is used to train the gear reducer fault samples,so that the FPN parameter values gradually approach the true value.The results of application examples show that this new algorithm for single or multiple fault diagnosis of gear reducer is very effective,the fault diagnosis accuracy has improved significantly,and the effectivity and correctness of optimization algorithm are verified.
作者
沙永东
侯静
徐广明
SHA Yongdong;HOU Jing;XU Guangming(Mechanical Engineering College,Liaoning Technical University,Fuxin 123000,China)
出处
《辽宁工程技术大学学报（自然科学版）》
CAS
北大核心
2019年第5期430-434,共5页
Journal of Liaoning Technical University (Natural Science)
基金
国家自然科学基金(51774161)
辽宁省教育厅高等学校青年科研项目(LJ2017QL018).
关键词
齿轮减速器
故障诊断
神经网络
FPN
网络参数
gear reducer
fault diagnosis
neural network
fuzzy petri nets
network parameters
