基于粒子群算法和变分模态分解的起重机滚动轴承性能退化评估

Performance degradation assessment of crane rolling bearing based on particle swarm optimization and variational mode decomposition

针对港口起重机滚动轴承早期故障特征不易提取、识别精度不够高的缺点,提出一种以粒子群(PSO)优化变分模态分解(VMD),结合支持向量数据描述(SVDD)的滚动轴承性能退化评估模型。通过PSO优化VMD中的参数,更好地提取滚动轴承的特征。利用SVDD模型中球心距离度量性能退化程度,并借助隶属度函数量化轴承性能退化,进而实现对轴承性能退化程度的精确评估。应用滚动轴承的全寿命试验数据验证该模型,并与传统的时域特征指标比较,本方法对轴承性能退化评估具有更强的敏感性,验证该方法的优越性。

A novel approach is proposed to assess the degradation of rolling bearing performance in harbor cranes.By integrating particle swarm optimization(PSO)with variational mode decomposition(VMD)and support vector data description(SVDD),the model addresses challenges in early fault characteristic extraction and identification accuracy.PSO optimizes VMD parameters(K andα)for improved feature extraction.The SVDD model employs the distance from the hypersphere center to quantify performance degradation and uses a membership function to assess bearing degradation accurately.Model validation is performed with full-life experimental data,comparing favorably against traditional time-domain indicators.