Adaptive fuzzy identification and optimization of CO2 capture capacity of CaO-based sorbents via sonochemical method in calcium looping

This study presents a novel approach that integrates adaptive fuzzy (AF) logic identification techniques with a genetic algorithm (GA) to optimize the synthesis of CaO sorbent from starch-modified limestone via a sonochemical method, aiming to enhance the CO2 capture capacity in the calcium looping cycle. In this study, adaptive fuzzy logic using the gradient descent method is employed to adjust the width and center of membership functions (MFs). Then, the AF identification model is used to inverse the output and determine the sonication input parameters. Integrating adaptive fuzzy logic with GA presents a novel approach for developing a robust model that accurately estimates the optimal parameters for the sonication process, significantly improving the CO2 capture capacity of CaO sorbents. The proposed model was also utilized to predict CO2 capture capacity and validated against experimental values. Additionally, its performance was compared with RSM and fixed fuzzy/GA, with error analysis performance criteria. The adaptive fuzzy GA demonstrated superior predictive accuracy and the capability for inverse modeling, enabling the determination of optimal input values required to achieve a desired CO2 capture capacity. This study highlights the potential of adaptive fuzzy GA integration in both predictive CO2 capture capacity and input parameters optimization, offering a pioneering methodology for enhancing CaO sorbent for CO2 capture applications.