Blade Imbalance Intelligent Diagnosis for wind Turbine Rotors Based on a Flexible Sensing System and a Deep Learning Method

Early detection of rotor imbalance is a prospective method for addressing wind turbine safety. A novel intelligent method based on a flexible sensing system and deep learning is proposed for diagnosing mass imbalance in rotor blades. A flexible sensing system that contains MEMS accelerometers, signal conditioning, and wireless transmission is integrated with the rotor blade for in-situ signal acquisition. Teager energy operator (TEO) is applied to demodulate and enhance the pulse compositions in vibration signals, resulting in Teager energy spectrums for model input. To fully extract the signal features, bidirectional temporal convolutional network (BiTCN) models are established to learn both the forward and backward signal features. Experimental verification shows that the proposed method detects the existence of imbalance with 100% accuracy, and the accuracies of localization and quantization are 98.26% and 97.57%, respectively, which are higher than those of the models with other input forms. In addition, compared with the other three different algorithms, BiTCN is superior in terms of prediction accuracy.