Francisco Alberto Castillo Leonardo

Plasmonic photocatalysis offers promising alternatives for solar energy conversion, hydrogen production, CO₂ reduction, and pollutant degradation. However, its performance depends on multiple variables related to composition, structure, optical and electronic properties, charge transfer, operating conditions, and reactor configuration. This systematic review analyzes the application of artificial intelligence (AI) and machine learning in photocatalysis, with an emphasis on plasmonic materials, during the period 2020–2026. Using a PRISMA-based methodology, supplemented by bibliometric and technical–comparative analyses, we examined material design and selection, the prediction of photocatalytic performance, reactor optimization, and the construction of datasets. The results show that AI models enable the exploration of compositional and synthetic spaces, the prediction of optical and electronic properties, the optimization of heterojunctions and core–shell structures, and the estimation of H₂ production, CO₂ conversion, and pollutant degradation, in addition to reducing the computational cost of simulations through surrogate models. Integration with DFT calculations, CFD simulations, active learning, and inverse design expands its application to interconnected predictive and experimental systems. However, limitations persist regarding small and heterogeneous datasets, insufficient external validation, interpretability, and low transferability. Advances in the field will require standardized and traceable data, physically meaningful descriptors, and models that bridge the material, reaction, and reactor scales.

This paper presents a systematic review conducted under the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) framework, analyzing 286 scientific articles focused on vibration-based predictive maintenance strategies for wind turbines within the context of advanced Prognostics and Health Management (PHM). The review combines international standards (ISO 10816, ISO 13373, and IEC 61400) with recent developments in sensing technologies, including piezoelectric accelerometers, microelectromechanical systems (MEMS), and fiber Bragg grating (FBG) sensors. Classical signal processing techniques, such as the Fast Fourier Transform (FFT) and wavelet-based methods, are identified as key preprocessing tools for feature extraction prior to the application of machine-learning-based diagnostic algorithms. Special emphasis is placed on machine learning and deep learning techniques, including Support Vector Machines (SVM), Random Forest (RF), Convolutional Neural Networks (CNN), Long Short-Term Memory networks (LSTM), and autoencoders, as well as on hybrid digital twin architectures that enable accurate Remaining Useful Life (RUL) estimation and support autonomous decision-making processes. The bibliometric and case study analysis covering the period 2020–2025 reveals a strong shift toward multisource data fusion—integrating vibration, acoustic, temperature, and Supervisory Control and Data Acquisition (SCADA) data—and the adoption of cloud-based platforms for real-time monitoring, particularly in offshore wind farms where physical accessibility is constrained. The results indicate that vibration-based predictive maintenance strategies can reduce operation and maintenance costs by more than 20%, extend component service life by up to threefold, and achieve turbine availability levels between 95% and 98%. These outcomes confirm that vibration-driven PHM frameworks represent a fundamental pillar for the development of smart, sustainable, and resilient next-generation wind energy systems.