This paper addresses multistage distribution network expansion planning (DNEP) incorporating energy storage systems (ESSs). The ESSs are utilized to shave the peak demand and to reduce the planning cost. Annual and daily load duration curves are considered to evaluate the impacts of the ESSs on the planning. The proposed planning is carried out based on the AC power flow including active power, reactive power, and network loss. The problem is formulated as a constrained, mixed-integer, and nonlinear programming (MINLP) and solved by using particle swarm optimization (PSO) algorithm. A 11 kV and 30-bus radial distribution network is considered as case study and the typical ESSs are also regarded to install on the network. Simulation results demonstrate the effectiveness and viability of the proposed method to consider the ESSs in DNEP. The results indicate that integrating the ESSs in DNEP reduces the planning cost significantly, as well improves the technical parameters of the network such as bus voltages and line loading. (C) 2015 Elsevier Ltd. All rights reserved.