• CDI different cell architectures are thoroughly reviewed. • Role of activated carbon electrodes in CDI performance is presented. • Machine learning model development for predictive CDI optimization. • Key challenges, limitations and future research directions for ML-assisted CDI were discussed. . Capacitive deionization (CDI) is an emerging and energy-efficient electrochemical water desalination technology, particularly for low-salinity sources. The performance of CDI systems is strongly influenced by both cell architecture and electrode physicochemical properties. This review systematically examines various CDI cell configurations, including membrane-less asymmetric and membrane-based symmetric and asymmetric systems, and evaluates their fundamental characteristics, removal performance, and energy demands. In parallel, it explores the role of activated carbon (AC) electrodes, such as specific surface area and specific capacitance, in affecting overall desalination efficiency, by varying CDI process parameters. Recent advances in machine learning (ML) have introduced powerful tools for predictive modelling and process optimization in CDI systems. Supervised learning models and ensemble techniques have shown potential in forecasting key performance indicators, including salt adsorption capacity, based on material and process parameters. This review assesses the current state of ML integration in CDI systems by utilizing the data from published articles. By combining insights from electrochemical engineering and data-driven modelling, this work outlines pathways toward intelligent, adaptive desalination systems. It concludes by proposing research directions that emphasize reproducibility, open data, and interdisciplinary approaches to advance ML-driven CDI for smart water infrastructure. .
Hai et al. (Sun,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: