An optimization method for the location and capacity of Electric Vehicle (EV) charging stations based on density peaks clustering of traffic flow is proposed to tackle the issue of EV demand. First, planning area traffic flow and parking lot opening index are analyzed, and the spatial distribution of EV charging demand data point collection is constructed. Then, the density peaks clustering method is used to analyze the density of the spatial distribution of charging demand, and the candidate clusters are obtained. Then, considering the cohesion and separation of clusters, the overall average contour coefficient is used to optimize the clustering results, so as to determine the location of the clustering center for EV charging stations. Further, based on the penetration rate of EVs, the number of EVs is predicted to determine the total charging demand of various types of EVs in the planned area. The charging station capacity of the corresponding cluster center is determined according to the charging demand proportion of each cluster. The location and capacity of EV charging stations in a particular urban area are determined, and the demonstration shows the feasibility of the method. This work is supported by the National Key Research and Development Program of China (No. 2017YFB0902800) and the Science and Technology Project of State Grid Henan Electric Power Company (No. SGHAYJ00GHJS1900031).