Brushless DC Motor Fault Diagnosis on the Basis of Complementary Fusion of Multiple-Channel Signals

Brushless DC motor is one of the important power devices for large-scale equipment, and the operating state of this motor is highly consistent with the operating state of the equipment. Existing motor fault diagnosis methods often target a single signal for feature extraction and diagnosis, and making accurate state judgment for motors in complex environments is difficult. To achieve reliable fault identification and classification, this study proposes a brushless DC motor diagnostic framework that leverages complementary fusion across multiple-channel signals. First, the fault signals are processed by the proposed fused Gramian angular difference field method to obtain a feature-enhanced image dataset. Second, a brushless DC motor fault diagnosis method called SimAM-ConvNeXt is developed to classify the fault image dataset and recognize different fault types. Last, a brushless DC motor fault diagnosis platform is built to collect fault signals in a real environment and construct a brushless DC motor fault diagnosis dataset. Experimental results demonstrate that the proposed method achieves a recognition accuracy of 99.8% on the electromechanical equipment fault diagnosis dataset, outperforming other approaches. The proposed approach is benchmarked against existing signal processing and fault recognition techniques to validate its superiority and effectiveness.