In order to increase the accommodation rate of renewable energy and cut the system operation cost, this paper takes “source-load-energy storage” coordination as an important mean to improve the system operation economy and the accommodation rate of renewable energy. Then, an optimization dispatch model based on “source-load-energy storage” coordination which considers thermal units, energy storage device and demand response as dispatching resources is established. The objective of this model including the operation costs of thermal units, the costs of demand response and the benefits from renewable energy consumption. Based on the different characteristics of the decision variables, a two-stage optimal algorithm is proposed. The first stage algorithm uses binary particle swarm optimization algorithm to minimize the start costs of thermal units, costs of wind power and photovoltaic curtailment and costs of demand response; the second stage uses double layer continuous particle swarm optimization algorithm to minimize the fuel costs of thermal units. The simulation results verify the validity of the proposed optimization dispatch strategy.