Because traditional control methods can no longer cope with the impact of high-permeability distributed generation on the power grid, the dynamic reconfiguration of a distribution network with distributed generation needs to be studied. There are problems of insufficient consideration of distributed generation, a complex and time-consuming process and low practicability in traditional reconfiguration. Therefore a two-stage optimization strategy for dynamic reconfiguration based on biogeography-based optimization is proposed, and a multi-objective optimization model is established to minimize both the power loss during the whole period and the total number of switching operations. First, an integer ring network coding method is used to reduce the dimension of variables, the distribution network is preliminarily divided into time periods, and the optimal interval is "enumerated" using the pinch strategy. All this can obtain a preliminary optimization scheme. The total number of switch operation constraints are considered. Then the secondary optimization of the time period is carried out, and the switch action timing and combination of dynamic reconfiguration are finally determined. An example shows that the dynamic reconfiguration method can reduce the active power loss of the distribution network, and improve the voltage stability of the nodes, while ensuring low operation times.