摘要
针对矿用电动机轴承在复杂工况下故障特征提取困难,提出一种基于改进型斑翠鸟优化算法(IPKO)优化连续变分模态分解(SVMD)的轴承故障特征提取方法。引入多策略融合改进方法,提升IPKO的收敛精度与全局寻优能力。构建以最小峭度为目标的适应度函数,将优化后的带宽约束参数α用于SVMD对振动信号进行模态分解。结合包络熵与峭度融合指标筛选最佳本征模态函数(IMF),并进行包络谱分析。与变分模态分解(VMD)对比结果表明,该方法在轴承故障特征提取中具有明显的优势。
Aiming at the difficulty of fault feature extraction of mining motor bearings under complex conditions,proposed a bearing fault feature extraction method based on Improved Pied Kingfisher Optimization(IPKO)optimizing Successive Variational Mode Decomposition(SVMD).A multi-strategy fusion enhancement method was introduced to improve IPKO convergence accuracy and global search ability.Built the fitness function taking minimum kurtosis as target and used the optimized bandwidth parameter α in SVMD for the mode decomposition of the vibration signal.Envelope entropy and kurtosis fusion indices were combined to select the most informative Intrinsic Mode Function(IMF)for envelope spectrum analysis.Comparative results with Variational Mode Decomposition(VMD)confirm this method′s superior performance in bearing fault feature extraction.
作者
孙霞
李文清
孙强
Sun Xia;Li Wenqing;Sun Qiang(School of Electrical and Information Engineering,Anhui University of Science&Technology,Huainan 232001,China)
出处
《煤矿机械》
2026年第3期173-178,共6页
Coal Mine Machinery
基金
国家自然科学基金项目(51874010)
安徽省质量工程项目(2020xsxxkc142)。
