Transformer reliability is crucial to grid security, with core loosening a common fault. This paper proposes a transformer core loosening fault diagnosis method based on a fusion feature extraction approach and Categorical Boosting (CatBoost) optimized by the Crested Porcupine Optimizer (CPO) algorithm. Firstly, the audio signal is decomposed into six Intrinsic Mode Functions (IMF) components through Variational Mode Decomposition (VMD). This paper utilizes Gaussian membership functions to quantify the energy proportion, central frequency, and kurtosis of IMF and constructs a fuzzy entropy discrimination function. Then, the IMF noise components are removed through an adaptive threshold. Subsequently, the denoised signal undergoes a wavelet packet transform instead of a short-time Fourier transform to optimize Mel-frequency cepstral coefficients (WPT-MFCC), combining time-domain statistical features and frequency-band energy distribution to form a 24-dimensional fusion feature. Finally, the CatBoost algorithm is employed to validate the effects of different feature schemes. The CPO is introduced to optimize its iteration number, learning rate, tree depth, and random strength parameters, thereby enhancing overall performance. The CPO-optimized CatBoost model had 99.0196% fault recognition accuracy in experimental testing, 15% better than the standard CatBoost. Accuracy exceeded 90% even under extreme 0 dB noise. This method makes fault diagnosis more accurate and reliable.

Aiming at the difficulty of fault feature extraction and low fault accuracy of rolling bearing, a method of optimized variational mode decomposition (VMD) based on the improved African vulture algorithm was proposed to extract bearing fault features efficiently. First, the Gaussian random walk and adaptive parameter optimization algorithm for African vultures (GAVOA) is introduced into the algorithm. Second, GAVOA is used to optimize the penalty factor and modal component of VMD. This algorithm can effectively avoid the local optimal solution and improve the search efficiency. Finally, the feature vector is used as the input of CNN‐BiLSTM to identify the fault types of rolling bearings. The experimental results show that the GAVOA‐VMD‐CNN‐BiLSTM fault diagnosis model proposed in this paper has better performance, and the fault diagnosis accuracy can reach 99.667%.