Answers are grounded in this paper and cite their sources.
Demonstrates improved energy production prediction in renewable systems, highlighting the effectiveness of a hybrid model.
Accurate energy production prediction is a critical challenge in modern renewable energy systems due to the inherent variability and complexity of influencing factors. This study proposes an optimized hybrid deep learning framework that integrates Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) enhanced by Genetic Algorithm (GA)-based hyperparameter optimization to improve predictive performance. A publicly available renewable energy dataset comprising 15,000 records and 13 features was utilized, where Z-score normalization was applied to standardize input variables and enhance model convergence. To reduce dimensionality and retain the most informative attributes, Binary Particle Swarm Optimization (BPSO) was employed for feature selection and benchmarked against Binary Grey Wolf Optimization (BGWO), Binary Whale Optimization Algorithm (BWOA), and Binary Butterfly Optimization Algorithm (BBOA), demonstrating superior performance in terms of fitness and error metrics. The dataset was partitioned into training (70%), validation (20%), and testing (10%) subsets to ensure robust generalization. The proposed CNN-RNN model was further optimized using GA and compared with several baseline regression models, including Gradient Boosting Regressor (GBR), Extra Trees Regressor (ETR), Decision Tree Regressor (DTR), K-Nearest Neighbors Regressor (KNNR), and Dummy Regressor (DR). Experimental results indicate that the optimized GA-CNN-RNN model significantly outperformed all competing models, achieving the lowest Mean Squared Error (MSE = 0.0098), Mean Absolute Error (MAE = 0.0791), Median Absolute Error (MedAE = 0.0683), Mean Absolute Percentage Error (MAPE = 0.0090), and the highest coefficient of determination (R 2 = 99.85%). Comparative analysis with alternative metaheuristic optimizers, including Artificial Bee Colony (ABC), Firefly Algorithm (FA), and Dragonfly Algorithm (DA), further confirmed the superiority of GA in enhancing model performance.
No takes yet. Share an insight, caveat, or question.
Alaa A. Almelibari (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: