为实现船舶电气故障的早发现、早解决，设计基于机器学习算法的船舶电气故障分类与诊断方法。采用Trager能量算子增强传感器采集到的船舶电气设备振动信号，利用小波包分析方法提取增强后的电气设备振动信号特征，将电气设备振动信号特征输入卷积神经网络中进行训练，得出最佳的故障分类与诊断模型，并利用该模型实现船舶电气设备的故障分类与诊断。实验表明：采用Teager能量算子可以快速准确地将传感器采集的信号放大，且放大过程没有信息损失。训练后卷积神经网络的故障分类与诊断正确率接近100%，可能够准确诊断出船舶电气设备是否存在故障，并获取对应的电气故障类型。
To achieve early detection and resolution of ship electrical faults, a machine learning algorithm based classification and diagnosis method for ship electrical faults is designed. The Trager energy operator is used to enhance the vibration signal of the electrical equipment of the ship collected by the sensor, and the wavelet packet analysis method is used to extract the characteristics of the enhanced vibration signal of the electrical equipment. The characteristics of the electrical equipment vibration signal are input into the convolutional neural network for training, and the best fault classification and diagnosis model is obtained. The experiment shows that using Teager energy operator can quickly and accurately amplify the signal collected by the sensor, and there is no information loss during the amplification process. After training, the accuracy of fault classification and diagnosis of convolutional neural network is close to 100%, which can accurately diagnose whether there is a fault in the ship's electrical equipment and obtain the corresponding electrical fault type.
2023,45(15): 143-146 收稿日期：2023-03-05
DOI：10.3404/j.issn.1672-7649.2023.15.029
分类号：TM315
基金项目：工信部2030型长江干线绿色智能船舶关键技术及示范船研制项目（CBG4N21-4-2）
作者简介：洪祥(1988-),男,硕士,高级工程师,主要研究方向为电力调度控制等
参考文献：
[1] 赵柄锡, 冀大伟, 袁奇, 等. 采用时域与时频域联合特征空间的转子系统碰磨故障诊断[J]. 西安交通大学学报, 2020, 54(1): 75–84
ZHAO Bing-xi, JI Da-wei, YUAN Qi, et al. Rubbing fault diagnosis of rotor system based on combined feature space in time and time-frequency domains[J]. Journal of Xi'an Jiaotong University, 2020, 54(1): 75–84
[2] 王瑞涵, 陈辉, 管聪, 等. 基于图卷积网络的非均衡数据船舶柴油机故障诊断[J]. 中国舰船研究, 2022, 17(5): 289–300
WANG Rui-han, CHEN Hui; GUANG Cong, et al. Fault diagnosis of marine diesel engines based on graph convolutional network under unbalanced datasets[J]. Chinese Journal of Ship Research, 2022, 17(5): 289–300
[3] 刘国强, 林叶锦, 张志政, 等. 基于粗糙集和优化DAG-SVM的船舶主机故障诊断方法[J]. 中国舰船研究, 2020, 15(1): 68–73
LIU Guo-qiang, LIN Ye-jin, ZHANG Zhi-zheng, et al. Main marine engine fault diagnosis method based on rough set theory and optimized DAG-SVM[J]. Chinese Journal of Ship Research, 2020, 15(1): 68–73
[4] 徐鹏, 杨海燕, 程宁, 等. 基于优化BP神经网络的船舶动力系统故障诊断[J]. 中国舰船研究, 2021, 16(S1): 106–113
XU Peng, YANG Hai-yan, CHENG Ning, et al. Fault diagnosis of ship power system based on optimized BP neural network[J]. Chinese Journal of Ship Research, 2021, 16(S1): 106–113
[5] 裴迪, 岳建海, 焦静. 基于自相关与能量算子增强的滚动轴承微弱故障特征提取[J]. 振动与冲击, 2021, 40(11): 101–108+123
PEI Di, YUE Jian-hai, JIAO Jing. Weak fault feature extraction of rolling bearing based on autocorrelation and energy operator enhancement[J]. Journal of Vibration and Shock, 2021, 40(11): 101–108+123
[6] 陈长征, 魏巍. 基于改进LMD与小波包降噪对故障弱信号的提取[J]. 机械设计与制造, 2020(347): 165–168+172
CHEN Chang-zheng, WEI Wei. Extraction of weak fault signal based on improved LMD and wavelet packet De-nosing[J]. Machinery Design & Manufacture, 2020(347): 165–168+172
