通过多源数据融合进行感应电机故障诊断的增强CNN
Likang Bo1, Meng Li1, Hua Tan1
1Hebei Vocational University of Technology and Engineering, Xingtai, Hebei , China.
PloS one
|August 22, 2025
概括
这项研究使用多源信息和卷积神经网络 (CNN) 增强了诱导电机故障诊断. 这种新方法在识别单一故障信号方面取得了很高的准确性,提高了电机的可靠性.
科学领域:
- 电气工程
- 机器学习
- 信号处理
背景情况:
- 感应电机的单一故障信号隔离因识别精度低而面临挑战.
- 错误特征的非线性相关性使传统的诊断方法变得复杂.
研究的目的:
- 为感应电机提出一种新的故障诊断策略,整合多源信息和增强的卷积神经网络 (CNN).
- 通过减轻非线性相关性来提高单个故障信号的识别精度.
主要方法:
- 使用无声自动编码器预处理振动和定位器电流信号.
- 使用相关差异贡献率方法将多个来源的同质数据合并.
- 合并的信号被转化为二维图像, 进入精致的CNN架构.
主要成果:
- 拟议的自适应CNN模型实现了高诊断精度:99.0%在1800r/min和94.8%在2400r/min.
- 该方法有效地利用来自多个来源的冗余和互补信息.
- 与传统方法和其他先进的多源聚变技术相比,其表现优越.
结论:
- 综合多源信息和CNN方法为诱导电机故障诊断提供了强大的工具.
- 该方法成功模拟非线性动态,并增强特征提取以提高准确性.
- 这一策略显著提高了感应电机故障检测系统的可靠性和精度.
更多相关视频
06:08A Cognitive Fusion-guided Prostate Biopsy Using Multiparametric Magnetic Resonance Imaging and Transrectal Ultrasound
Published on: March 21, 2025
374
07:12Using Informational Connectivity to Measure the Synchronous Emergence of fMRI Multi-voxel Information Across Time
Published on: July 1, 2014
12.4K
相关概念视频
Three-Phase Short Circuit—Unloaded Synchronous Machine
235
Conducting a three-phase short circuit test on an unloaded synchronous machine helps understand its impact on the system. The AC fault current's oscillogram, with the DC offset removed, reveals that the waveform amplitude decreases from an initially high value to a steady-state level for one phase of the machine.
This behavior occurs due to the magnetic flux produced by the short-circuit armature currents. Initially, these currents follow high-reluctance paths but eventually shift to...
This behavior occurs due to the magnetic flux produced by the short-circuit armature currents. Initially, these currents follow high-reluctance paths but eventually shift to...
235
Power System Three-Phase Short Circuits
148
Determining the subtransient fault current in a power system involves representing transformers by their leakage reactances, transmission lines by their equivalent series reactances, and synchronous machines as constant voltage sources behind their subtransient reactances. In this analysis, certain elements are excluded, such as winding resistances, series resistances, shunt admittances, delta-Y phase shifts, armature resistance, saturation, saliency, non-rotating impedance loads, and small...
148
