Multi‐component condition monitoring method for wind turbine gearbox based on adaptive noise reduction

Wind turbine gearbox condition monitoring (C&M) is a key technology to promote wind farm maintenance cost reduction and power generation improvement. Existing C&M methods usually adopt the component-by-component modelling approach. Firstly, this approach inefficient in modelling; secondly, due thermal conduction effect, abnormalities one component affect other components, making it difficult identify source of faults. To solve these problems, normal behaviour model (NBM) combining data adaptive noise an improved variational auto-encoder (VAE) proposed, which can monitor operational multiple components simultaneously takes into account correlation between when warning specific abnormal component. As verified by practical cases, method balances accuracy efficiency multi-component NBM achieves effective early accurate localization abnormalities. The proposed has lower false alarm missed rates compared single-component NBMs.