Key Takeaways & Executive Findings
- •• • On rainy days, the proposed model reduces ERMSE and EMAE by 49.3% and 51.9% on average compared to a single BiLSTM model, directly improving forecast reliability under volatile weather—critical for grid frequency regulation and reducing reserve capacity requirements. • • The entropy-based causal feature extraction method quantifies nonlinear coupling among meteorological factors, eliminating redundant inputs that plague linear correlation methods like Pearson coefficients, thereby enhancing feature selection quality and model generalization. • • The parallel BiLSTM and TCN-Transformer architecture captures multi-scale temporal dependencies: BiLSTM handles short-term fluctuations, while TCN-Transformer extracts local patterns and long-range dependencies, achieving superior performance over single models in all three weather scenarios. • • The improved Q-learning algorithm with a cosine-modulated dynamic reward-penalty mechanism adaptively adjusts sub-model weights, overcoming the static weight limitations of error inverse and mean fusion methods, and preventing premature convergence or local optima during training.
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
Photovoltaic (PV) power forecasting is challenged by redundant meteorological features, insufficient multi-timescale dynamic response, and the lack of dynamic adaptability in hybrid models. This study proposes an ultra-short-term forecasting framework integrating causal feature extraction with an improved Q-learning (IQ-L) algorithm. First, an entropy-based causal feature extraction method quantifies nonlinear coupling between meteorological variables and PV power, filtering key factors. Second, a parallel prediction model combining bidirectional long short-term memory (BiLSTM) and TCN-Transformer captures short-term fluctuations, local patterns, and long-term dependencies, enabling multi-scale feature fusion. Third, an improved Q-learning algorithm with a dynamic reward-penalty mechanism, incorporating a cosine factor, iteratively adjusts the weights of sub-models to produce the final forecast. Validation under three typical weather conditions shows that the proposed model achieves stable performance. On rainy days, compared with a single BiLSTM model, the proposed method reduces ERMSE and EMAE by an average of 49.3% and 51.9%, respectively, demonstrating enhanced accuracy and generalization. The framework addresses the limitations of static weight allocation and single reward mechanisms in existing hybrid approaches, offering a robust solution for grid dispatch and energy storage optimization.
1. Introduction
Existing commercial PV forecasting solutions predominantly rely on static weight allocation or simple ensemble methods, such as error inverse weighting or mean fusion. These approaches are inherently vulnerable to outliers and lack dynamic responsiveness to rapidly changing meteorological conditions. Furthermore, traditional feature extraction techniques—like Pearson correlation—only capture linear relationships, failing to quantify the nonlinear coupling between meteorological variables and PV power output. This limitation leads to redundant inputs, increased model complexity, and degraded prediction accuracy, particularly under highly variable weather such as rain or cloud cover. Consequently, grid operators face challenges in scheduling and energy storage optimization, as forecast errors propagate into dispatch decisions.
To address these bottlenecks, this study introduces a hybrid forecasting framework that integrates causal feature extraction with an improved Q-learning algorithm. The entropy-based causal method filters key meteorological factors by quantifying nonlinear interactions, reducing input dimensionality while preserving critical information. A parallel BiLSTM and TCN-Transformer model captures multi-scale temporal features, and an improved Q-learning algorithm with a cosine-modulated dynamic reward-penalty mechanism adaptively adjusts the contribution of each sub-model. This protocol directly tackles the static weight and single reward limitations of prior work, delivering a 49.3% and 51.9% reduction in ERMSE and EMAE on rainy days compared to a single BiLSTM, thereby enhancing forecast stability and generalization for real-world deployment.
Loading authentic research manuscript (Pages 1–5)...
ZHANG Li, LIU Jiawei, SUN Shuyan, ZHANG Tao, ZHANG Hongwei (2026). Ultra-Short-Term Photovoltaic Power Forecasting Based on Causal Feature Extraction and an Improved Q-Learning Algorithm. Acta Energiae Solaris Sinica. https://doi.org/10.19912/j.0254-0096.tynxb.202608_9677
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 hybrid forecasting models does the proposed IQ-L algorithm address?
Existing hybrid models often use static weight allocation (e.g., error inverse method) that is sensitive to outliers and lacks dynamic adjustment, leading to degraded accuracy during rapid weather transitions. Additionally, single reward mechanisms in Q-learning can cause insufficient exploration in early training or premature convergence to local optima. The proposed IQ-L algorithm introduces a cosine-modulated dynamic reward-penalty mechanism that adaptively balances exploration and exploitation, and iteratively adjusts sub-model weights based on prediction error feedback, thereby mitigating these failure modes.
How does the entropy-based causal feature extraction improve upon traditional linear correlation methods like Pearson coefficients?
Pearson coefficients only capture linear relationships and cannot quantify nonlinear coupling among multiple meteorological variables. The entropy-based causal method computes causal entropy to measure nonlinear interactions and indirect influences, enabling the selection of key meteorological factors that are strongly causally linked to PV power fluctuations. This reduces feature redundancy and enhances the model's ability to generalize under complex weather conditions, as evidenced by improved performance across three typical weather scenarios.
What are the scalability and computational bottlenecks when deploying this model for real-time ultra-short-term forecasting?
The parallel BiLSTM and TCN-Transformer architecture increases computational load compared to single models, potentially challenging real-time deployment on edge devices. However, the causal feature extraction reduces input dimensionality, partially offsetting this. The IQ-L algorithm requires iterative training, but once trained, inference is fast. For scalability, the model can be parallelized on GPUs, and the dynamic weight adjustment is lightweight. Field validation on a 13-month dataset (as per the paper's experimental setup) demonstrates stable performance, but further optimization is needed for resource-constrained environments.
How does the model perform under extreme weather conditions, such as heavy rain or rapid cloud transients, compared to benchmark models?
Under rainy conditions, the proposed model reduces ERMSE and EMAE by 49.3% and 51.9% on average compared to a single BiLSTM model. This improvement is attributed to the multi-scale feature fusion and dynamic weight adjustment, which better capture rapid fluctuations. The TCN-Transformer component excels at extracting local patterns and long-term dependencies, while the IQ-L algorithm adapts weights in real time. However, the paper does not provide specific metrics for extreme events like thunderstorms; further testing under such scenarios is recommended.
What is the economic impact of the reported accuracy improvement on grid dispatch and energy storage optimization?
A 49.3% reduction in ERMSE and 51.9% in EMAE directly lowers forecast uncertainty, enabling more accurate scheduling of reserve capacity and reducing reliance on fast-ramping fossil fuel plants. For energy storage, improved forecasts optimize charge-discharge cycles, extending battery life and reducing operational costs. While the paper does not provide a detailed cost-benefit analysis, the accuracy gains translate to fewer penalties for deviation in electricity markets and improved integration of PV into the grid, potentially saving millions in balancing costs annually for large-scale deployments.
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.