Key Takeaways & Executive Findings
- •• • GADF transformation converts 1D vibration signals into 2D images with sparse matrix encoding, preserving temporal correlation and suppressing noise; this directly addresses the <90% accuracy ceiling of conventional EMD/Teager-based methods and enables reliable early-stage fault warning in wind farms. • • Embedding CBAM in the CNN convolutional layers provides dual channel and spatial attention, increasing global feature extraction; this overcomes the shallow two-layer convolution limitation of prior multiscale CNNs and improves discrimination of weak fault signatures under variable load. • • Replacing ReLU with the βc-ACONC activation function eliminates neuron necrosis and enables selective activation, preserving valid features while filtering noise; this extends the model's usable training depth and stabilizes convergence on high-dimensional GADF inputs. • • ISSA-optimized XGBoost hyperparameters deliver >99% diagnostic accuracy on the laboratory gearbox dataset, a decisive improvement over the 20–30% fault contribution and 10–15% maintenance cost burden of gearbox failures, enabling condition-based maintenance with direct cost reduction.
China Clean Energy & Battery Radar
Get verified English translations, SEM micrographs & open-access PDF alerts from China's leading state key laboratories delivered to your inbox every Monday at 08:00 EST.
Abstract
Gearbox failures account for 20–30% of total wind turbine faults and incur maintenance costs equivalent to 10–15% of overall turbine value. Conventional vibration diagnostic pipelines—complementary ensemble empirical mode decomposition with singular value energy spectrum, time-varying filtering empirical mode decomposition, and Teager energy spectrum analysis—remain bounded below 90% accuracy and depend on expert-driven feature engineering that is sensitive to non-stationary operating conditions and noise. This study proposes an intelligent diagnostic architecture that converts one-dimensional gearbox vibration signals into two-dimensional images via Gramian angular difference field (GADF) transformation, preserving intrinsic temporal correlation and time-frequency structure while exploiting matrix sparsity to suppress interference. An improved convolutional neural network (CNN) extracts multi-dimensional features: a convolutional block attention module (CBAM) is embedded in the convolutional layers to weight critical channels and focus on fault-sensitive spatial regions, and a modified βc-ACONC activation function replaces ReLU to mitigate neuron necrosis and enable selective activation. The extracted composite features are then fed into an XGBoost network whose hyperparameters are optimized by an improved sparrow search algorithm (ISSA). Validation on a laboratory wind turbine gearbox dataset yields diagnostic accuracy exceeding 99%, demonstrating robust fault identification capability under complex operating conditions.
1. Introduction
Gearbox failures represent 20–30% of all wind turbine faults and drive maintenance expenditure equal to 10–15% of total turbine value. High torque, variable loading, and environmental shock produce tooth surface wear and tooth breakage that demand early detection. Vibration signals remain the primary diagnostic data source because they directly reflect gear meshing state and damage signatures. Yet established techniques—complementary ensemble empirical mode decomposition with singular value energy spectrum, time-varying filtering empirical mode decomposition, and Teager energy spectrum analysis—extract only partial time-frequency features, rely on expert experience and manual feature engineering, and remain sensitive to non-stationary signals and noise. Their diagnostic accuracy typically falls below 90%. Wavelet analysis captures transient features but suffers from difficult basis function selection and poor generalization.
Deep learning offers an end-to-end alternative, but existing implementations remain constrained. Short-time Fourier transform with CNN-SVM improves classification accuracy yet still operates on one-dimensional signals. Variational mode decomposition with African vulture optimization reduces feature extraction complexity but does not resolve non-stationarity. CEEMDAN-CNN-BiLSTM enhances temporal modeling but inherits the inherent limitations of 1D representations. Multiscale CNNs use only two convolutional layers, leaving deep features underextracted. Deep belief networks with locust optimization are training-intensive and struggle with high-dimensional data. One-dimensional signals fundamentally weaken the mapping between fault features and fault types due to limited time-frequency resolution and spatial representation. This study addresses the bottleneck by converting vibration signals into GADF images that preserve time-frequency correlation and suppress noise through matrix sparsity, then applying an improved CNN with CBAM and βc-ACONC activation, and finally optimizing an XGBoost classifier via ISSA to achieve >99% diagnostic accuracy.
Loading authentic research manuscript (Pages 1–5)...
WANG Yan, WANG Zijian, ZHONG Xinqi, LIANG Shiyu, ZHAO Hongshan (2026). Wind Turbine Gearbox Fault Diagnosis Method Based on Improved CNN-XGBoost Fusion Model Under Gramian Angular Difference Field. Acta Energiae Solaris Sinica. https://doi.org/10.19912/j.0254-0096.tynxb.202608_9733
Research & Educational Purpose Only: The translations, structured abstracts, analytical annotations, and data reports provided by SinoGreenTechare intended exclusively for academic research, internal corporate R&D, and educational benchmarking. They do not constitute formal engineering, chemical safety, legal, or professional advice.
Copyright & Intellectual Property Notice: Original copyright of the underlying source articles and experimental data remains with the respective authors, institutions, and original publishing journals. SinoGreenTech claims intellectual property only over its proprietary translations, analytical syntheses, and AEO structured enhancements in accordance with international fair use and academic citation principles.
Frequently Asked Questions
What specific failure mechanisms in wind turbine gearboxes does the GADF-CNN-XGBoost model target, and how does it overcome the non-stationarity that defeats conventional EMD-based methods?
The model targets tooth surface wear and tooth breakage driven by high torque, variable load, and environmental shock. Conventional EMD-based methods—complementary ensemble empirical mode decomposition with singular value energy spectrum, time-varying filtering EMD, and Teager energy spectrum—are sensitive to non-stationary signals and noise, yielding diagnostic accuracy below 90%. GADF transformation maps 1D vibration signals into 2D images via polar coordinate encoding, preserving temporal correlation and nonlinear dynamic features. The resulting matrix sparsity suppresses interference and enhances the identifiability of weak amplitude differences, enabling the CNN to extract fault-sensitive features that conventional methods miss.
How does the βc-ACONC activation function improve upon ReLU in this CNN architecture, and what is the measurable impact on training stability and feature representation?
ReLU induces neuron necrosis, which halts learning in affected units and weakens feature representation. The βc-ACONC function introduces a selective activation mechanism that dynamically adjusts channel nonlinear responses, preserving valid features while filtering noise. This mitigates neuron necrosis, maintains gradient flow, and improves the network's overall feature expression capability. The result is stable convergence on high-dimensional GADF inputs and a diagnostic accuracy exceeding 99% on the laboratory gearbox dataset, compared to the sub-90% ceiling of conventional pipelines.
What is the industrial cost-benefit of achieving >99% diagnostic accuracy compared to the 20–30% fault contribution and 10–15% maintenance cost burden of gearbox failures?
Gearbox faults account for 20–30% of total wind turbine failures and maintenance costs equal 10–15% of overall turbine value. A diagnostic accuracy above 99% enables early fault warning and condition-based maintenance, directly reducing unplanned downtime and repair expenditure. The model's ability to identify weak fault signatures under complex operating conditions allows operators to schedule maintenance before catastrophic tooth breakage occurs, avoiding secondary damage to the drivetrain and extending gearbox service life.
How does the ISSA-optimized XGBoost classifier compare to standalone CNN or XGBoost approaches in terms of hyperparameter tuning and generalization on high-dimensional GADF features?
Standalone CNNs suffer from insufficient global feature extraction and ReLU-induced neuron necrosis, while unoptimized XGBoost is sensitive to hyperparameter selection and prone to overfitting on high-dimensional inputs. The improved sparrow search algorithm (ISSA) optimizes XGBoost hyperparameters, and the CNN with CBAM extracts composite features that capture both channel and spatial dependencies. This fusion architecture achieves >99% diagnostic accuracy on the laboratory wind turbine gearbox dataset, demonstrating superior generalization compared to single-model baselines and prior deep learning methods such as CEEMDAN-CNN-BiLSTM and multiscale CNNs.
What are the scalability and deployment bottlenecks for implementing this GADF-CNN-XGBoost model on operational wind farms with heterogeneous gearbox fleets and varying sensor configurations?
The model requires vibration signal acquisition consistent with the laboratory dataset used for validation. GADF transformation is computationally lightweight and preserves signal integrity, while the CNN-XGBoost fusion is trained end-to-end. Deployment bottlenecks include sensor placement consistency, sampling rate uniformity, and the need for retraining when gearbox geometry or operating conditions differ substantially. The ISSA optimization adds offline computational cost but does not affect inference latency. The >99% accuracy achieved on the laboratory dataset provides a baseline; field validation across heterogeneous fleets remains necessary to confirm robustness under real-world noise and load variability.
Related Chinese Research & Cross-Citations
Joint Forecasting of Wind and Photovoltaic Power Considering Complementarity
The inherent spatiotemporal complementarity between wind and solar resources offers a theoretical basis for improving renewable power forecasting accuracy. This study proposes a joint wind-photovoltaic (PV) power forecasting strategy that explicitly exploits this complementarity. A bidirectional long short-term memory (BiLSTM) neural network serves as the baseline forecasting model, and a novel sorting and comparative optimization (SCO) algorithm is developed to optimize the model's hyperparameters. The SCO algorithm ranks individuals in ascending order and compares adjacent fitness values to escape local optima, a known deficiency in conventional metaheuristics such as genetic algorithms and particle swarm optimization. For wind farms and PV plants exhibiting significant complementarity, the joint forecasting strategy first aggregates their power outputs, normalizes the combined signal, and then feeds it into the optimized BiLSTM model. Experimental results demonstrate that the proposed SCO-BiLSTM model reduces the eMAPE by 10.313% compared with PSO-BiLSTM. Furthermore, joint forecasting under SCO-BiLSTM lowers the eRMSE by 27.443% relative to standalone PV power forecasting. The study also establishes that forecasting accuracy improves with stronger wind-solar complementarity but degrades as the forecasting horizon extends. These findings confirm that exploiting complementarity in joint forecasting substantially enhances predictive performance for renewable energy integration.
Ultra-Short-Term Wind Power Forecasting Based on Fluctuation Continuation Scenario Identification
Existing ultra-short-term wind power forecasting methods exhibit limited performance due to insufficient extraction of fluctuation information and inadequate analysis of evolution patterns. This paper proposes an ultra-short-term wind power forecasting method based on fluctuation continuation scenario identification. First, the coupling mechanism of wind power fluctuations under multiple turbulence processes is investigated, and historical wind power dynamics are decoupled into a combination of nonlinear and linear fluctuation components. A fluctuation continuation concept is introduced, and the future continuation scale of wind power fluctuations is derived from nonlinear and linear decoupling parameters, thereby classifying fluctuation continuation scenarios. A sparse neural network (SNN) oriented to high-dimensional sparse features is constructed to identify historical fluctuation continuation scenarios, and ultra-short-term power forecasting is conducted separately for each scenario. Validation using measured wind speed and power data from three wind farms shows that, compared with baseline models, the proposed model improves root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) by at least 1.46%, 2.44%, and 14.67%, respectively, demonstrating superior accuracy and stability. The method addresses the limitations of signal decomposition techniques that lack physical interpretability and are sensitive to hyperparameters, and overcomes the high-dimensional sparsity challenges faced by traditional scenario identification models.
Multi-Classifier Open Adversarial Network for Rolling Bearing Fault Diagnosis in Wind Turbine Generator Systems
Rolling bearings in wind turbine generator systems operate under variable speed and load conditions that induce significant data distribution shifts between training and field data, while unknown fault modes absent from the source domain are frequently misclassified as known classes. This study proposes a multi-classifier open adversarial network (MCOAN) for open-set fault diagnosis. Within an adversarial domain adaptation framework, a K-way classifier and an additional K+1-way one-vs-all classifier independently estimate target-sample similarity to the source domain. These similarity scores drive a dynamic weighting mechanism that adaptively reweights target samples during open-set adversarial training and supplies per-sample dynamic thresholds for known/unknown discrimination, thereby promoting cross-domain alignment of shared-class features while suppressing negative transfer from unknown samples. A non-adversarial domain classifier is introduced to stabilize dynamic weight estimation. Validation on two datasets demonstrates high-precision shared-class distribution alignment and unknown-class recognition with favorable robustness. The method removes reliance on empirically preset thresholds that plague conventional open-set back-propagation approaches, where a fixed threshold of 0.5 in the binary cross-entropy adversarial loss provides no per-sample adaptivity. By coupling K-way and K+1-way similarity estimates, MCOAN achieves simultaneous known-class alignment and unknown-class separation without prior knowledge of the unknown-class cardinality, addressing a persistent bottleneck in wind turbine drivetrain condition monitoring where unanticipated bearing failure modes emerge under field conditions.
Unsupervised Automated Identification Method for Abnormal States of Wind Turbine Gearboxes
Addressing the scarcity of labeled data for training classification models in wind turbine planetary gearbox anomaly identification, this study proposes an unsupervised automated detection method. Log Mel-band energy features are extracted from raw vibration signals and fed into an unsupervised anomaly recognition model centered on a U-net autoencoder. A health-state threshold is established based on reconstruction error between model input and output, enabling anomaly identification. The method is validated using factory gearbox test data and operational data from a wind farm in Yangtouya, Shanxi. For factory gearboxes, dual validation is performed using a spectrum amplitude modulation-based signal processing method. Results demonstrate that the proposed method achieves 93.34% recognition accuracy on both factory and wind farm test sets, confirming its capability to automatically and correctly separate abnormal wind turbine gearboxes. The approach eliminates reliance on labeled fault data, offering a scalable solution for full-lifecycle health monitoring, from factory acceptance testing to in-service early anomaly detection, adaptable across different operating conditions and turbine models.
Improved Adaptive Super-Twisting Sliding Mode Control for Permanent Magnet Synchronous Motors
Adaptive super-twisting sliding mode control (ASTSMC) for permanent magnet synchronous motors (PMSM) suffers from prolonged convergence and insufficient disturbance rejection under complex operating conditions. This paper proposes an improved ASTSMC incorporating a fixed-time disturbance observer (FTDO). A power term is introduced into the adaptive super-twisting controller to accelerate convergence far from the origin, while the discontinuous sign function is replaced by a continuous h(s) function to mitigate chattering. The FTDO ensures disturbance estimation converges within a fixed time independent of initial states, overcoming the limitations of traditional and finite-time observers. The estimated disturbance is fed forward to the sliding mode controller for compensation, enhancing robustness. The FTDO design is based on an auxiliary state variable z and its derivative, with error dynamics analyzed via Lyapunov stability. Comparative simulations against conventional disturbance observers and sliding mode controllers validate the proposed strategy. The results demonstrate shorter convergence time, improved dynamic performance, and superior disturbance rejection, making the approach suitable for high-performance servo systems. The method addresses the critical need for robust, fast-response control in electric vehicles, rail transit, and aerospace applications where PMSM drives face significant uncertainties and external disturbances.
Data-Model Jointly Driven Fault Diagnosis for Wind Turbine Planetary Gearboxes
Fault diagnosis of wind turbine planetary gearboxes is severely constrained by the scarcity of high-quality fault data, as gearboxes operate predominantly in healthy states and automatic shutdowns prevent fault progression. This paper proposes a data-model jointly driven diagnosis method to address low diagnostic accuracy under limited fault samples. A high-fidelity lumped-parameter dynamic model of the planetary gearbox is constructed to generate pseudo-fault data, supplementing the training set. A domain-shared residual network feature extractor incorporating convolutional block attention modules extracts key physical features from both pseudo and measured data. Local maximum mean discrepancy aligns feature distributions at the fault-category level between pseudo and real fault data. A Kolmogorov-Arnold network module enhances the model's capacity to learn complex data relationships, enabling classification and identification of different fault types. Validation on a wind turbine planetary gearbox fault diagnosis test rig demonstrates that the proposed method achieves superior diagnostic performance under fault sample scarcity compared to classical methods. The framework offers an effective solution for known fault types, though identification of unknown and atypical faults remains a challenge for future work via open-set domain generalization.