Wind turbine planetary gearboxes have complex structures and operating environments, which makes it difficult to extract fault features effectively. In addition, it is difficult to achieve efficient fault diagnosis. To improve the efficiency of feature extraction and fault diagnosis, a fault diagnosis method based on improved multivariate variational mode decomposition (IMVMD) and ensemble refined composite multivariate multiscale dispersion entropy (ERCmvMDE) with multi-channel vibration data is proposed. Firstly, the IMVMD is proposed to obtain the optimal parameters of the MVMD, which would make the MVMD more effective. Secondly, the ERCmvMDE is proposed to extract rich and effective feature information. Finally, the fault diagnosis of the planetary gearbox is achieved using the least squares support vector machine (LSSVM) with features consisting of ERCmvMDE. Simulations and experimental studies indicate that the proposed method performs feature extraction well and obtains higher fault diagnosis accuracy.

Efficient extraction and classification of fault features remain critical challenges in planetary gearbox fault diagnosis. A fault diagnosis framework is proposed that integrates hierarchical refined composite multiscale fuzzy entropy (HRCMFE) for feature extraction and a gray wolf optimization (GWO)-optimized least squares support vector machine (LSSVM) for classification. Firstly, the HRCMFE is developed for feature extraction, which combines the segmentation advantage of hierarchical entropy (HE) and the computational stability advantage of refined composite multiscale fuzzy entropy (RCMFE). Secondly, the hyperparameters of LSSVM are optimized by GWO using a proposed fitness function. Finally, fault diagnosis of the planetary gearbox is achieved by the optimized LSSVM using the HRCMFE-extracted features. Simulation and experimental study results indicate that the proposed method demonstrates superior effectiveness in both feature discriminability and diagnosis accuracy.

The vibration signal of rolling beating is nonlinear and nonstationary, which makes feature extraction difficult for fault diagnosis. To improve the efficiency of feature extraction and fault diagnosis, a hybrid model based on optimized variational mode decomposition (VMD), fuzzy dispersion entropy (FDE), and a support vector machine (SVM) is proposed. Firstly, a parameter optimization method using the sparrow search algorithm (SSA) was applied to VMD to improve the decomposition ability. Subsequently, a feature vector based on the FDE was proposed as a fault‐diagnosis feature. Finally, SVM was applied with the proposed feature vector for the fault diagnosis of rolling bearings. The simulation and experimental study results indicate that the proposed method can obtain useful features for fault diagnosis, particularly in short‐length samples and noise conditions. The proposed method performed well in the fault diagnosis for different fault types and degrees of rolling bearings.

To address the limited performance of transfer fault diagnosis for planetary gearboxes under cross-operating conditions, which is caused by the heterogeneous feature distribution of vibration data and insufficient feature extraction. An improved domain-adversarial neural network (IDANN) model based on a joint-adaptive-domain alignment component and a dual-branch feature extractor is proposed. Firstly, a joint domain adaptation alignment approach, integrating maximum mean discrepancy (MMD) and CORrelation ALignment (CORAL), is proposed to realize the correlation structure matching of features between the source and target domains of IDANN. Secondly, a dual-branch feature extractor composed of ResNet18 and Swin Transformer is proposed with an attention-weighted fusion mechanism to enhance feature extraction. Finally, validation experiments conducted on public planetary gearbox fault datasets show that the proposed method attains high accuracy and stable performance in cross-operating-condition transfer fault diagnosis.

Data distribution shift across different rotational speeds limits domain generalized fault diagnosis for planetary gearboxes. Existing single-representation domain generalization (DG) methods degrade severely in cross-speed extrapolation, with different representations showing speed-dependent advantages. The core contribution of this paper is a sample-wise confidence-adaptive decision fusion mechanism under the pure DG constraint. For each test sample, the fusion weights are computed online from the relative softmax prediction confidence of the two branches. The mechanism requires neither target-domain data nor calibration sets and introduces no fusion hyperparameters. A physics-guided dual-branch representation is constructed to support this mechanism. The physics-guided order-domain feature (PGF) branch extracts speed-independent features via triple-band demodulation and computed order tracking resampling. Physical enhancement is further applied to improve cross-domain robustness. A homologous time-domain branch shares the IEDGNet architecture and training strategy. Experiments on the WT planetary gearbox dataset show that the proposed method achieves 92.23% average accuracy in forward speed extrapolation, which is 5.62 percentage points higher than the best single branch and only 0.72 percentage points lower than the global Oracle upper bound, and reaches 96.48% accuracy in backward generalization. Complementarity analysis and feature visualization reveal non-overlapping misclassification regions between branches, confirming that the confidence arbitration mechanism effectively exploits dual-representation complementarity in cross-speed extrapolation scenarios.

Vibration signals of planetary gearboxes under variable-speed conditions exhibit strong non-stationarity and modulation, so a single feature representation cannot comprehensively describe intricate fault patterns, and distribution discrepancies across rotating speeds degrade the cross-condition generalization of diagnostic models. To address these limitations, this paper proposes an adaptive domain-aligned multi-modal feature fusion network (ADAMFFN). Three parallel branches extract complementary features from dual-channel vibration signals: spatial coupling features from orbit images, time–frequency energy features from continuous wavelet transform (CWT) representations, and frequency-domain statistical (FreqStat) features from power and envelope spectra. Heterogeneous features are mapped into a shared latent subspace through a unified projection layer, deep cross-modal interaction is realized by a progressive fusion network, and a domain alignment mechanism based on a domain-adversarial neural network (DANN) is introduced to eliminate source–target distribution gaps via adversarial training. On eight leave-one-speed-out (LOSO) cross-speed tasks constructed on the public WT-Planetary Gearbox dataset, ADAMFFN achieves an average accuracy of 99.25%, outperforming the best single-branch and dual-branch schemes by 2.63 and 0.40 percentage points, respectively; ablation experiments verify the complementarity of the three modalities and the effectiveness of domain alignment. Cross-condition external validation on the Southeast University (SEU) gearbox dataset further demonstrates its generalization capability under a different test rig and acquisition conditions.