This work introduces a novel approach based on utilizing intelligent data-based optimization techniques to fabricate high-quality and low-cost carbon electrodes for capacitive deionization (CDI). The effects of synthesis parameters of peanut shell-derived porous carbon (PSPC) (KOH to Raw ratio, pyrolysis temperature and time) on five factors (production yield, BET surface area, electrode specific capacitance, and ion transfer resistances, including interface and diffusion) were investigated using combined design-of-experiment (DOE), responsesurface-methodology (RSM), and Sobol’s sensitivity analysis (SA). Optimal scenarios providing maximum production yield, surface area, and specific capacitance, and minimum ion transfer resistances were found using a multi-objective genetic algorithm (GA). Results uncover that increasing the impregnation ratio, pyrolysis temperature and time enhances the PSPC porosity that causes the carbon surface area and electrode specific capacitance to promote, the production yield to reduce and ion transfer resistances to raise. From Sobol’s SA, pyrolysis temperature was the most impacting parameter on all decision factors. The optimal PSPC prepared at 750 ◦C and minimum KOH:Raw = 1 and time = 30 min offered a developed mesoporous structure with BET surface area of 685 m2 /g and the resulting electrode had an outstanding specific capacitance (127.13 at 5 mV/s), with reliable durability after 100 cycles and showing the lowest possible ion transfer resistances.
