This paper presents the utilization of polyethylenimine (PEI) polymer to obtain freestanding, binder-free, and nitrogen- enriched carbon nanofibers (NCNFs) by electrospinning which is an easy and fast method. The PEI polymer played a significant role as a poring agent for the adjustment of surface area and as the heteroatom doping to increase conductivity. NCNFs-10 electrode exhibited a specific capacitance of 200 F/g at 1 A/g with a good retention of 84.25 % at 10 A/g in three electrode system. The assembled supercapacitor had a specific energy of 5.57 Wh/kg at 125 W/kg with a good cycling performance of 80.5 % after 10,000 cycles. Furthermore, an artificial neural network (ANN) method was utilized to understand the relationship between the CNF's structure and supercapacitor performance by modeling the specific capacitance. The input factors were specific surface area (S_BET), nitrogen doping (N% at), graphitization degree (I_D/I_G), internal resistance (R_s), pore size (V_por), micropore volume (V_mic), the average diameter (D_ave), and mesopore volume (V_meso). Results showed that the first three remarkable variables are S_BET, N (%at), and I_D/I_G.

本文介绍了利用聚乙烯亚胺（PEI）聚合物通过静电纺丝获得独立式、无粘合剂、富氮碳纳米纤维（NCNF），这是一种简单、快速的方法。PEI聚合物作为调节表面积的成孔剂和作为提高电导率的杂原子掺杂发挥了重要作用。NCNFs-10 电极在三电极系统中在 1 A/g 电流下表现出 200 F/g 的比电容，在 10 A/g 电流下具有 84.25% 的良好保留率。组装的超级电容器在125 W/kg下的比能量为5.57 Wh/kg，10,000次循环后具有80.5%的良好循环性能。此外，人工神经网络(ANN) 方法通过对比电容进行建模来了解 CNF 结构与超级电容器性能之间的关系。输入因素为比表面积（S _BET）、氮掺杂（N% at）、石墨化程度（ID_/IG_）、内阻（R_s）、孔径（V_por）、微孔体积（V_mic） 、平均直径 (D_ave) 和介孔体积 (V_meso)。结果显示，前三个显着变量是 S_BET、N (%at) 和 I_D/I_G。