基于频域自注意力编码的起重机回转支承寿命预测方法  

Service life prediction method of crane slewing bearing based on frequency domain self-attention coding

作　　者：顾永华 苏文胜[1,2] 薛志钢 曾汉生[1,2] 许晨旭 Gu Yonghua;Su Wensheng;Xue Zhigang;Zeng Hansheng;Xu Chenxu

机构地区：[1]江苏省特种设备安全监督检验研究院,南京210036 [2]国家桥门式起重运输机械产品质量检验检测中心,无锡214174

出　　处：《起重运输机械》2024年第20期61-66,共6页Hoisting and Conveying Machinery

摘　　要：起重机回转支承的使用寿命预测对提高设备维护效率、降低运营成本及确保生产安全至关重要。文中提出了基于频域自注意力编码的起重机回转支承寿命预测方法,旨在解决传统预测方法精度和效率上的局限。利用深度学习中的自注意力机制将起重机回转支承的振动信号从时域转换到频域,并通过自注意力编码提取频域中的主导特征,实现对起重机回转支承剩余使用寿命的精确预测。该方法不仅有效降低了模型的计算复杂度,提高了预测精度和效率,而且通过实际实验验证了其在起重机回转支承寿命预测领域的良好应用潜力和实用价值。同时,将频域分析与自注意力机制相结合,为起重机回转支承寿命预测提供了新的解决思路。The service life prediction of crane slewing bearing is very important to improve equipment maintenance efficiency,reduce operating costs and ensure operational safety.In this paper,a service life prediction method of slewing bearing based on frequency domain self-attention coding was proposed to eliminate limitations of traditional prediction methods in accuracy and efficiency.The self-attention mechanism in deep learning was adopted to transform the vibration signal of slewing bearing from time domain to frequency domain,and the dominant features in frequency domain were extracted by self-attention coding,so as to accurately predict the remaining service life of slewing bearing.This method can effectively reduce the computational complexity of the model,improve the prediction accuracy and efficiency,and it has been proved that it has a bright application prospect and practical value in the field of crane slewing bearing service life prediction.Additionally,the combination of frequency domain analysis and self-attention mechanism provides a new idea for the service life prediction of crane slewing bearing.

关 键 词：起重机 回转支承 寿命预测 自注意力机制 频域 

分 类 号：TH213.4[机械工程—机械制造及自动化]