摘要
针对港口起重机电机在频繁启停、高负载冲击及海洋环境侵蚀下易发的故障问题,提出基于振动信号分析的电机故障检测方法。首先对港口起重机电机的结构特点及常见故障类型进行归纳,明确了振动信号在机械故障诊断中的关键作用;其次设计了涵盖传感器选型、安装位置、采集系统搭建与信号预处理的完整实验流程;再次构建了包含监督学习与深度学习两大类的故障诊断模型,并通过网格搜索与贝叶斯优化等策略进行超参数调优;最后基于实际试验平台和多工况故障数据集,采用准确率、精确率、召回率、F1值与AUC五项指标对模型性能进行了全面评估。实验结果表明,所提方法在多种故障类型识别上均可实现超过97%的分类准确率,为港口起重机电机在线监测与智能维护提供了可靠支持。
In response to the frequent faults of port crane motors caused by frequent starts and stops,high-load impacts,and marine environmental corrosion,a fault detection method for motors based on vibration signal analysis is proposed.Firstly,the structural characteristics and common fault types of port crane motors are summarized,emphasizing the crucial role of vibration signals in mechanical fault diagnosis.Secondly,a comprehensive experimental procedure is designed,covering sensor selection,installation positions,acquisition system setup,and signal preprocessing.Thirdly,fault diagnosis models are constructed,incorporating both supervised learning and deep learning approaches,with hyperparameter tuning performed through strategies such as grid search and Bayesian optimization.Finally,based on an actual test platform and a multi-condition fault dataset,the model performance is comprehensively evaluated using five metrics:accuracy,precision,recall,F1-score,and AUC.The experimental results demonstrate that the proposed method achieves a classification accuracy exceeding 97%for various fault types,providing reliable support for online monitoring and intelligent maintenance of port crane motors.
作者
邵志成
SHAO Zhicheng(Shanghai Zhenhua Heavy Industries Co.,Ltd.,Shanghai 200125,China)
出处
《电工技术》
2025年第S1期1-3,共3页
Electric Engineering
