Global sensitivity analysis and optimization for coal–water slurry preheaters based on metamodel of optimal prognosis

Enhancing energy efficiency by coal–water slurry preheaters in the gasification process, the geometry of ladder-type fold helical baffles and shell–inlet velocity have an important effect on thermo-hydraulic performance of preheaters. A variance-based global sensitivity analysis procedure was built by metamodel of prognosis, and optimization results were compared between genetic algorithm and particle swarm optimization. The simulation results illustrate that heat transfer coefficient per unit pressure drop (K/ΔP) increases when folding angle and folding ratio increase, and K/ΔP decreases with increasing relative height and shell–inlet velocity. As for K/ΔP, the coefficient of prognosis of full model is 99.03%, and total effects of shell–inlet velocity, folding angle, folding ratio and relative height is 70.34%, 25.01%, 8.74%, and 7.87%, respectively. Based on reduced spaces with most influential parameters, optimization cases show genetic algorithm is more efficient and provides better results, folding angle, folding ratio, relative height and shell–inlet velocity are 50°, 0.6, 0.5 and 1.4 m s^–1 when maximizing K/ΔP. The research results provide theoretical guidance for selection, design and optimization of coal–water slurry preheaters.