Fig. 1: Schematic representation of the hybrid models and the application domains. | Nature Energy

Fig. 1: Schematic representation of the hybrid models and the application domains.

From: A hybrid modelling approach to compare chemical separation technologies in terms of energy consumption and carbon dioxide emissions

Fig. 1

a, Schematic representation of this work: hybrid modelling for the chemical separation industry. The curated NF-10K dataset was used in the machine learning and energy models to forecast the applicability of nanofiltration in the untapped chemical space (that is, unexplored separations). The eight implications and results of the hybrid modelling are expected to aid industrial separation design and decision-making. Eevaporation, energy of evaporation; Enanofiltration, energy of nanofiltration; Eextraction, energy of extraction; S, solvent; R, rejection; logP, log partition coefficient. b, The three chemical separation technologies considered in this research: triple-effect continuous evaporator, continuous nanofiltration and liquid–liquid extraction. Details of assumptions and considerations are given in Supplementary Methods.

Back to article page