Key Takeaways & Executive Findings
- •• • The proposed T-CGAN achieves superior performance in generating 8760-hour annual PV output scenarios, overcoming the gradient vanishing and mode collapse issues that plague LSTM-GAN and SeqGAN when directly trained on long sequences; this enables reliable long-term power system planning and annual operational scheduling without the boundary feature loss and cross-month correlation degradation inherent in short-sequence stitching approaches. • • The multi-scale periodic attention mechanism, incorporating a learnable periodic bias matrix P[i,j] = cos(2π|i-j|/T) with adjustable strength parameter β, effectively captures both diurnal (short-term) and seasonal (long-term) periodic patterns in PV output; this is critical for accurately representing the deterministic periodic components that dominate PV generation variability across annual cycles. • • The Wasserstein distance with gradient penalty as the adversarial loss function ensures stable training convergence and mitigates mode collapse, a common failure mode in standard GAN training for time-series generation; this stability is essential for industrial deployment where model retraining and reproducibility are required. • • Comparative experiments against three typical models (WGAN-GP, LSTM-GAN, and SeqGAN) demonstrate the proposed method's superiority across statistical metrics, temporal correlation, and scenario validity, confirming that Transformer-based architectures with periodic attention are better suited for long-duration renewable energy scenario generation than recurrent or convolutional alternatives.
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
The inherent intermittency and volatility of photovoltaic (PV) power pose significant uncertainties for power system planning and operation. Existing generative adversarial network (GAN)-based scenario generation methods predominantly focus on short-term horizons and suffer from gradient vanishing, mode collapse, and boundary feature loss when extended to long time scales (e.g., annual 8760-hour sequences). This paper proposes a Transformer-CGAN (T-CGAN) model that integrates an improved Transformer architecture with a conditional GAN. Both generator and discriminator employ Transformer-based neural networks, with seasonal labels incorporated as conditional information. A multi-scale periodic attention mechanism is developed to capture diurnal and seasonal periodic features of PV output. The Wasserstein distance with gradient penalty is adopted as the loss function to enhance training convergence and stability. Validation using historical data from a PV station in northwestern China demonstrates that the proposed method effectively captures long- and short-term temporal dependencies, accurately generating annual PV output scenario sets. Comparative experiments against three typical models verify the method's superiority in statistical metrics, temporal correlation, and scenario validity.
1. Introduction
Existing commercial approaches to PV output scenario generation have largely relied on probabilistic sampling or short-horizon machine learning models. Sampling methods depend on prior distribution assumptions that fail to capture the complex nonlinear temporal characteristics of PV output. GAN-based methods such as WGAN, WGAN-GP, and LSTM-GAN have been applied to short-term scenarios but exhibit fundamental limitations when extended to annual 8760-hour horizons: recurrent architectures suffer from gradient vanishing or explosion, mode collapse, and computational inefficiency during direct training on long sequences. The conventional workaround—training on short sequences and stitching them together—introduces boundary feature loss, data discontinuities, and loss of cross-month long-term correlations, rendering the generated scenarios unsuitable for annual planning and long-term electricity market analysis.
The proposed T-CGAN addresses these bottlenecks by replacing recurrent components with a Transformer architecture that simultaneously attends to all positions in the sequence, capturing global dependencies without sequential processing constraints. A multi-scale periodic attention mechanism is introduced to explicitly model the diurnal and seasonal periodicities inherent in PV generation, while seasonal labels provide conditional guidance. The Wasserstein distance with gradient penalty stabilizes adversarial training. This protocol specifically targets the industrial need for accurate, high-fidelity annual PV output scenario sets that preserve both short-term fluctuations and long-term seasonal patterns, enabling robust power system planning and market trading decisions.
Loading authentic research manuscript (Pages 1–5)...
YE Yujiang, XUAN Shunde, SHI Ruifeng, JIA Limin (2026). Long-Time-Scale Photovoltaic Output Scenario Generation Method Based on Improved Transformer-CGAN. Acta Energiae Solaris Sinica. https://doi.org/10.19912/j.0254-0096.tynxb.202608_9675
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 existing GAN-based methods does the proposed T-CGAN overcome when generating annual 8760-hour PV output scenarios?
Existing methods such as TimeGAN, SeqGAN, and LSTM-GAN suffer from gradient vanishing or explosion, mode collapse, and low training efficiency when directly trained on long sequences. The common workaround of training on short sequences and stitching them together causes boundary feature loss, data mutation, and loss of cross-month long-term correlations. The T-CGAN replaces recurrent architectures with a Transformer that captures global dependencies via self-attention, incorporates a multi-scale periodic attention mechanism to explicitly model diurnal and seasonal periodicities, and employs Wasserstein distance with gradient penalty to stabilize adversarial training, thereby eliminating these failure modes.
How does the multi-scale periodic attention mechanism quantitatively improve the capture of PV output periodicity compared to standard self-attention?
The mechanism introduces a learnable periodic bias matrix P[i,j] = cos(2π|i-j|/T), where T is the period length and i,j are position indices, added to the standard self-attention logits with a learnable strength parameter β. This explicitly encodes both short-term (diurnal) and long-term (seasonal) periodic patterns. Standard self-attention lacks this inductive bias, requiring the model to learn periodicity solely from data, which is inefficient and prone to overfitting. The periodic bias ensures that the attention mechanism inherently favors periodic relationships, improving generalization and accuracy in capturing PV output's deterministic periodic components.
What is the industrial significance of using Wasserstein distance with gradient penalty as the loss function in T-CGAN for long-term PV scenario generation?
The Wasserstein distance with gradient penalty provides a smoother and more stable gradient signal compared to the original GAN minimax loss, which is notorious for training instability and mode collapse. In the context of long-term PV scenario generation, stable training is critical because the model must learn complex temporal dependencies over 8760 hours without diverging. This stability ensures reproducibility and reliability, which are prerequisites for industrial deployment in power system planning and electricity market analysis, where scenario sets must be generated consistently and accurately.
How does the T-CGAN's performance compare to existing models in terms of statistical metrics and temporal correlation for annual PV output scenarios?
The paper reports comparative experiments against three typical models (WGAN-GP, LSTM-GAN, and SeqGAN) using historical data from a PV station in northwestern China. The T-CGAN demonstrates superior performance across statistical metrics, temporal correlation, and scenario validity. Specifically, it effectively captures both long- and short-term temporal dependencies, accurately generating annual PV output scenario sets. The exact numerical improvements are detailed in the full experimental section, but the qualitative conclusion is that the Transformer-based architecture with periodic attention outperforms recurrent and convolutional alternatives for long-duration renewable energy scenario generation.
What are the scalability bottlenecks for deploying T-CGAN in real-world power system planning, and how does the architecture address them?
Scalability bottlenecks include computational complexity of self-attention (quadratic in sequence length) and memory requirements for processing 8760-hour sequences. The Transformer architecture processes sequences in parallel rather than sequentially, enabling efficient utilization of modern GPU hardware. The multi-scale periodic attention mechanism reduces the effective search space by incorporating periodic inductive biases, allowing the model to learn long-range dependencies with fewer parameters. Additionally, the conditional GAN framework with seasonal labels enables controlled generation, which is essential for scenario-based planning under different seasonal conditions. These features collectively address scalability for annual scenario generation.
Related Chinese Research & Cross-Citations
Wind Turbine Gearbox Fault Diagnosis Method Based on Improved CNN-XGBoost Fusion Model Under Gramian Angular Difference Field
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.
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.