摘要
当前,船舶低速柴油机常见机械故障处理效果不佳,因无法精准诊断机械故障类型,导致故障得不到有效检修,为此本文研究船舶低速柴油机常见机械故障诊断与检修策略。采用滑动平均滤波技术对离散振动信号平滑处理,并对其标准化处理;通过对原始振动信号短时傅里叶变换,将时域中信号转换为频域信息,得到低速柴油机振动频谱图;采用卷积神经网络对输入频谱图对应的机械故障状态诊断,根据诊断结果采取对应的对策,实现船舶低速柴油机常见机械故障诊断与检修。实验证明,能够实现对机械故障精准诊断,并且故障检修后柴油机运行状态趋于稳定。
Due to the poor application effect of the current method in the common mechanical faults of ship low speed diesel engines,the failure cannot be effectively repaired due to the inability to accurately diagnose the mechanical failure type,so the diagnosis and maintenance strategy of ship low speed diesel engines are proposed.The sliding average filtering technology is used to smooth the discrete vibration signal and standardize it;convert the signal to the low-speed diesel engine;the convolution neural network is used to diagnose the mechanical fault state corresponding to the input spectrum,and take the corresponding countermeasures according to the diagnosis results to realize the diagnosis and maintenance of common mechanical faults of low-speed diesel engine.The experiment proved that the design method can realize the accurate diagnosis of mechanical faults,and the operation state of the diesel engine tends to be stable after the fault maintenance,which can realize the effective diagnosis and maintenance of common mechanical faults of low-speed diesel engine.
作者
姜黎明
龙江林
张志勇
Jiang Li-ming;Long Jiang-lin;Zhang Zhi-yong(Offshore Oil Engineering Co.,Ltd.,Tianjin 30000,China)
出处
《内燃机与配件》
2025年第15期73-75,共3页
Internal Combustion Engine & Parts
关键词
柴油机
机械故障
检修
诊断
滑动平均滤波技术
短时傅里叶变换
Diesel engine
Mechanical failure
Maintenance
Diagnosis
Sliding average filtering technology
Short-time Fourier transform
