Channel-Spatial attention convolutional neural networks trained with adaptive learning rates for surface damage detection of wind turbine blades

•A novel channel-spatial attention convolutional neural network trained with adaptive learning rate is proposed for surface damage detection of wind turbine blades.•A channel-spatial attention scheme is introduced into WTB damage detection it assigns higher weights to damage-sensitive channels and enhances the spatial information of damage features.•A novel adaptive learning rate is designed, which enables the model to reach the global optimum faster and more accurately.