From Prediction to Process Design: A Systematic Review of AI/ML Integration with Chemical Engineering in Wastewater Treatment

Wastewater treatment is becoming increasingly complex due to variability in influent composition, the nonlinear response of treatment processes, and appearance of new contaminants. The customary empirical and mechanistic design practices often fail to capture these dynamics, resulting in systems that are expensive, rigid, and suboptimal. Artificial intelligence (AI) and machine learning (ML) offer forecasting, responsive, and data-driven solutions that can redesign processes. With the PRISMA ۲۰۲۰ framework, the review systematically evaluates scientific works published within ۲۰۱۵-۲۰۲۵ in bibliographic databases like Scopus, Web of Science, etc. Eligible works were filtered based on reproducibility, performance metrics, and scalability. The data shows that AI/ML approaches, such as artificial neural networks, fuzzy logic, and hybrid models, have been effectively utilized to streamline processes, anticipate contaminant removal, identify operational faults, and maximize energy use in a pilot as well as a full-scale plant. One of the unique contributions of this review is the introduction of AI/ML into the fundamental principles of chemical engineering, including kinetics, reactor design, mass transfer, and process control. Existing limitations include data quality, model interpretability, computational scalability, and regulatory acceptability. AIs/ML integration has the potential to enhance intelligent and more sustainable wastewater management, which is consistent with worldwide sustainable development goals.

کلیدواژه ها:

Artificial intelligence ، Digital twins ، Explainable AI ، Hybrid Modeling ، Machine Learning ، Wastewater Treatment