Photovoltaic power generation is a promising method of power generation that is not affected by geographical characteristics. However, it is difficult in urban areas to supply sufficient electricity to high-rise buildings through photovoltaic power generation owing to the limited roof area. As a countermeasure, photovoltaic power generation on flat wall surfaces has attracted attention and demonstrated its high potential. In this study, using GIS software, aerial photo LiDAR data, basic map information, and historical weather data, area-level solar power generation in several Japanese cities was simulated as power generation for a smart grid using a simulation system. In Japan ’s six locations (Chofu, Sapporo, Kyoto, Hachiken, Takasaki, and Oyama), power generation on the vertical surfaces of buildings exceeded horizontal surfaces power generation. The total solar power generation was expected to be twice as much as using the top surface only. Furthermore, a city close to the south in Japan obtained more solar power generation. Besides, power generation is strongly influenced by the urban form; power generation potential at the wall surface in residential areas is greater than in central areas. Based on the simulation results, solar radiation, latitude, and the characteristics of cities, a log-linear model to estimate solar power generation was proposed and evaluated. Wall photovoltaic power generation is an effective renewable energy source in a smart grid.