基于模拟退火算法的机车齿轮箱故障诊断系统
System Fault Diagnosis of Electric Locomotive Gear Box Based on Simulation Annealing
投稿时间:2012-12-11  
中文关键词:电力机车  齿轮箱  BP神经网络  模拟退火算法  故障诊断
英文关键词:electric locomotive  gear box  bp neural network  simulation annealing  fault diagnosis
基金项目:青藏铁路公司科委科技研究开发计划(QZ2010 J02)
作者单位
范万里 兰州交通大学 机电工程学院 
李刚 兰州交通大学 机电工程学院 
白宇君 兰州交通大学 机电工程学院 
高晓玲 兰州交通大学 机电工程学院 
摘要点击次数: 1263
全文下载次数: 1456
中文摘要:
      为了预测电力机车齿轮箱存在的潜伏性故障,设计了一种电力机车齿轮箱故障诊断系统。该系统通过VB调用Access数据库和基于模拟退火算法的BP神经网络模型,实现了对电力机车齿轮箱故障诊断的可视化操作。根据光铁谱技术原理,磨粒浓度的变化体现了齿轮箱内部的故障类型。因此,利用模拟退火的思想建立BP网络模型,收集电力机车齿轮箱的故障数据并进行归一化处理,处理后的数据作为网络的输入,故障类型的编码作为网络的目标输出。对模型进行了仿真和测试,结果表明该模型诊断电力机车齿轮箱故障准确率较高,可用于电力机车齿轮箱故障诊断系
英文摘要:
      The system fault diagnosis of electric locomotive gear box is designed in order to predict the potential fault of electric locomotive gear box, which realizes visualized operation of electric locomotive gear box fault diagnosis. The system calls the access database and the bp neural network model based on simulation annealing. According to the principle of RDEAES and Ferrography,the change of abrasive concentration reflects the type of gear box internal fault. Therefore,the bp network model is established to use the idea of simulation annealing, collect the fault data of electric locomotive gear box. The processed data act as the input of network and the codes of fault type act as the target output of network.The simulation and test of the model show that the model has a higher accuracy for electric locomotive gear box fault diagnosis and can be used in electric locomotive gear box fault diagnosis system.
范万里,李刚,白宇君,高晓玲.基于模拟退火算法的机车齿轮箱故障诊断系统[J].石家庄铁道大学学报:自然科学版,2013,(3):70-73.
查看全文  查看/发表评论  下载PDF阅读器
关闭