Condition　monitoring　and　fault　diagnosis　of　bearings　play　an　important　role　in　the　safe　operation　of　equipment　and　can　reduce　maintenance　costs.　In　this　paper,　a　novel　data-driven　bearing　fault　diagnosis　model　is　developed.　First,　the　variable　modal　decomposition　method　is　applied　for　denoising　and　recombination　to　reduce　noise　interference.　Next,　the　refined　composite　multi-scale　fuzzy　entropy　is　used　to　extract　features　from　the　recombined　signal.　After　that,　discriminant　diffusion　maps　analysis　is　utilized　to　compress　the　high-dimensional　features　into　the　low-dimensional　space　and　remove　the　interference　of　redundant　features.　Finally,　the　beetle　antennae　search　support　vector　machine　is　adopted　for　fault　classification.　The　proposed　method　is　applied　to　the　fault　diagnosis　of　wind　turbine　bearings　under　various　operating　conditions,　and　the　experimental　results　show　that　the　proposed　method　can　accurately　and　effectively　identify　various　faults.　©　2021　IEEE.