Based on the system state transition sampling approach，this paper presents a sequential Monte Carlo simulation method for the reliability evaluation of the distribution system including distributed generations（DGs）with high penetration rate.The proposed method can be used to calculate both the generation and the distribution reliability indices.The state transition models of non-power components and DGs are established respectively.The chronological output of wind turbines，photovoltaic arrays and the time varying load as well as the utility grid capacity are all taken into account.The state of charge of battery is simulated by the Kinetic Battery Model（KiBaM）incorporating different battery dispatch strategies.The state space of the system is divided into two categories of the generation states and the distribution states.The system is treated as a distribution system when some non-power component is in its down state，and treated as generation system when all the non-power components are in up states.The validity of the proposed method was demonstrated by a study case.This method can be used for the optimal planning of distribution networks including DGs.