Traditional simulation work often starts from the study of the impact of a single factor on device performance to obtain the optimal value of that factor and then regards the combination of the optimal values of each factor as the optimization condition. Obviously, this approach ignores the impacts of the interactions among factors on device performance. To address this issue, this paper uses Quokka3 v2.6.0 and JMP Pro 17.0.0 to perform device simulation and parameter optimization research on tunnel oxide passivated back contact (TBC) solar cells. First, Quokka3 was employed to investigate the effects of silicon wafer properties, rear-side passivation and contact characteristics, and rear-side geometry on the performance of TBC solar cells. Subsequently, a total of 625 simulations were performed by using Quokka3 with four factors (wafer thickness, wafer resistivity, P/N ratio, and pitch) at five levels. Finally, JMP Pro was used to analyze the simulation results statistically. It was found that the pitch, P/N ratio, quadratic power terms, quadratic interaction terms except the interaction between wafer thickness and resistivity, cubic power terms, and some cubic interaction terms all have significant impact on power conversion efficiency (PCE). JMP Pro predicted that the TBC solar cell could achieve the maximum PCE of 26.784% under the conditions of wafer thickness = 143.25 μm, wafer resistivity = 1.09 Ω·cm, P/N ratio = 1.94, and pitch = 380 μm.
Chen et al. (Wed,) studied this question.