This paper proposes a novel structure for distribution network reconfiguration (DNR) to improve reliability indices and decrease the operational cost of power systems. In order to increase the applicability of the problem in the smart grid, a distribution network involving plug-in hybrid electric vehicles (PHEV) and distributed wind generation (RDG) is addressed. The aim is to determine the optimized configuration of the distribution network considering different scenarios of PHEV’s charging/discharging and power realization of RDGs. To hedge against the nonlinearity of the problem, an improved genetic algorithm (IGA) is proposed. The suggested IGA decreases the time burden calculation of the problem, which is necessary for DNR problems. Moreover, a fast radial checking algorithm (RCA) is suggested to avoid tedious radial checking procedures. Numerical studies are implemented on IEEE 33-bus test system. The results show that the suggested approach optimizes the configuration of the distribution network when a renewable power shortage occurs or PHEVs’ operation is rescheduled.