基于1DCNN-LSTM 神经网络模型的闸门故障诊断分析
Gate Fault Diagnosis and Analysis Based on 1DCNN-LSTM Neural Network Model
魏庆镇 1张依潇 2张浩潍2
作者信息
- 1. 肥城市尚庄炉水库管理中心,山东肥城 271600
- 2. 西南交通大学信息科学与技术学院,四川成都 611700
- 折叠
摘要
针对闸门故障导致安全事故的问题,提出了一种1DCNN-LSTM故障诊断模型,该方法结合了空间特征提取与时序特征理解能力,能够更全面地理解信号中蕴含的特征信息,通过对闸门系统的荷载电流、开度等特征信号进行特征提取,更加高效地检测出闸门的工作状态,完成闸门故障诊断任务.结果表明:该方法的分类准确率达到了93.7％,且综合性能良好,相较于对比模型具有显著优势,充分证明了其有效性.
Abstract
Aiming at the problem of safety accidents caused by gate faults,a 1DCNN-LSTM fault diagnosis model is proposed.The proposed method combines the ability of spatial feature extraction and temporal feature to understand the feature information contained in the signal more comprehensively.The working state of the gate can be detected more efficiently,and the task of gate fault diagnosis can be completed.The results show that the classification accuracy of the proposed method reaches 93.7％,and the comprehensive performance is fine,which has significant advantages over the comparison model,and fully proves its effectiveness.
关键词
水工闸门/故障诊断/卷积网络/物联网Key words
Hydraulic gate/Fault diagnosis/Convolutional networks/Internet of Things引用本文复制引用
出版年
2024
NETL
NSTL
万方数据