一种轴承寿命预测方法及装置

• Format: Patent
• Language: Chinese
• Published: 23.07.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; MEASURING; PHYSICS; TESTING; TESTING STATIC OR DYNAMIC BALANCE OF MACHINES ORSTRUCTURES; TESTING STRUCTURES OR APPARATUS NOT OTHERWISE PROVIDED FOR

本发明公开了一种轴承寿命预测方法及装置，其中，方法包括以下步骤：步骤S1、获取轴承全寿命周期的振动信号，对振动信号进行预处理生成原始RRMS信号；步骤S2、建立复合神经网络作为特征提取网络，将原始RRMS信号输入特征提取网络，得到原始RRMS信号的物理特征；步骤S3、对PE模型进行参数确认，得到轴承的经验退化物理模型，将物理特征输入经验退化物理模型，得到重建RRMS信号；步骤S4、基于原始RRMS信号以及重建RRMS信号作为样本信号，并建立损失函数对神经网络进行训练，得到轴承寿命预测模型；步骤S5、获取待测轴承的待测RRMS信号，将待测RRMS信号输入轴承寿命预测模型，得到轴承寿命预测结果。本发明具有轴承寿命预测结果具有可解释性的效果。 The invention discloses a bearing life prediction method and device, and the method comprises the following steps: S1, obtaining a vibration signal of a whole life cycle of a bearing, and carrying out the preprocessing of the vibration signal, and generating an original RRMS signal; s2, establishing a composite neural network as a feature extraction network, and inputting the original RRMS signal into the feature extraction network to obtain a physical feature of the original RRMS signal; s3, performing parameter confirmation on the PE model to obtain an empirical degradation physical model of the bearing, and inputting the physical characteristics into the empirical degradation physical mode