Intelligent Energy-Efficient design and Multi-Objective optimization of recycle micromixers for sustainable Microchemical process intensification

Abstract

Microfluidic process intensification plays a key role in improving the sustainability and energy efficiency of advanced chemical systems by reducing resource consumption while enhancing transport phenomena. In this context, recycle-based micromixers offer superior mixing performance through induced flow recirculation, although their complex hydrodynamics introduce a strong trade-off between mixing quality and pressure-driven energy demand. This study proposes an intelligent, data-driven framework for energy-efficient design and multi-objective optimization of recycle micromixers in line with sustainable process development goals. The framework integrates computational fluid dynamics (CFD), gene expression programming (GEP) surrogate modeling, a multi-objective special relativity search (MOSRS) algorithm, and PROMETHEE-based decision analysis. A CFD dataset was generated by varying key geometric and operating parameters, including splitter length, splitter width, channel depth, number of mixing units, and Reynolds number. The optimization simultaneously maximizes mixing index and minimizes pressure drop as a proxy for pumping energy consumption. The GEP models achieved high predictive accuracy (R² > 0.999), enabling efficient surrogate-based optimization. The MOSRS algorithm generated a well-distributed Pareto front, highlighting the trade-off between high mixing efficiency (84–100 %) and low energy losses. Results indicate that higher Reynolds numbers and deeper channels enhance mixing, while shorter splitter lengths reduce hydraulic energy requirements.