Artificial intelligence and data-driven approaches in renewable energy: A review of achievements and challenges

As the global shift to sustainable energy accelerates, renewable energy systems (RES) increasingly depend on artificial intelligence (AI) and data-driven methods to improve predictive modeling, optimize operations, and integrate with power grids. This review looks at how AI and data methods help solve key renewable energy problems like unreliable supply, uncertainty, and difficult optimization. A clear review process was used. It focused on studies from 2020 to 2025 found in databases such as ScienceDirect, Scopus, and IEEE Xplore. Following PRISMA guidelines, about 150 core studies were chosen from many thousands for detailed review. Results show that neural networks, like CNNs, RNNs, LSTMs, and GNNs, are commonly used for forecasting and fault detection. Hybrid models that mix deep learning with optimization algorithms, such as genetic algorithms, work well for system design and energy management. Reinforcement learning and methods like random forests also help with real-time decisions for microgrids and energy storage. Also, physics-informed machine learning is accurate for studying battery aging, controlling load frequency, and improving electrolyzer cells. This review introduces a new multidimensional taxonomy that organizes AI applications by algorithm type, integration approach, system use, and policy considerations—addressing a gap in earlier reviews. Current trends highlight AI-enabled coordination across multiple energy systems and resilience planning enhanced by metaheuristics. While these technologies show strong potential, their widespread adoption faces obstacles such as inconsistent data quality, limited real-world testing, and adaptability across domains. To advance this field, the development of standardized benchmarks and advanced hybrid algorithms is recommended. This study maps a practical route for using AI and data-driven innovation to support the worldwide transition to renewable energy.