Lithium ion batteries (LiBs) are used in electric vehicles (EVs), hybrid electric vehicles (HEVs), compact devices and so on. However, it is required to have more power density and more energy density, so active materials, binders, electrolytes, and separators are enthusiastically developed [1, 2]. Moreover, the performance of LiB depends on not only materials but also the electrode structure. Because the optimal electrode structure is unknown and different by material characteristics, it is manufactured by trial and error. The purpose of this study is optimization of design parameters of the electrode structure from material characteristics by simulation. Moreover, to optimize design variables based on real structure, the simulation of porous structures is used. The purpose of optimization was to maximize the gravimetric energy density under power requirement constraints [3]. The design variables were volume ratio of active material, volume ratio of sub material (the mixture of carbon black and binder), volume ratio of void space filled by electrolyte, thickness of electrode layer and particle diameter of active material in cathode and anode, respectively. LiCoO2 and graphite were employed as cathode and anode active materials, respectively. To calculate an energy density, reaction and mass transport at the galvanostatic discharge were simulated, based on the porous electrode theory [4, 5]. In this work, the following assumptions were used; each composition in electrode layer was uniform, the temperature distribution in the cell was uniform and constant. To calculate an energy density, not only design variables but also parameters affected by electrode structure such as tortuosity are necessary, the method to determine these parameters is important. In this work, using the simulation of porous structure, structural parameters were calculated from design variables. This simulation first located active materials randomly at three dimensional space. Next, it located sub materials depending on pore size. Finally, tortuosity was calculated by random-walk method [6]. Because the calculation load of the way to simulate porous structure and calculate structural parameter in optimization was large, we adopted the way to calculate predictive equation prepared by simulation of porous structure in advance. Figure 1 shows the calculation results of optimal volume ratio of active material, volume ratio of sub material, volume ratio of electrolyte and thickness under each power requirement constraint. The difference of design variables of cathode between power requirements was remarkable as compared with one of design variables of anode. In cathode, an increase in power requirement caused a decrease in volume ratio of active material and an increase in volume ratio of electrolyte and thickness.