To ensure the dynamic reconfiguration of distribution network system safe and stable operation, a dynamic reconfiguration method for distribution network is proposed based on integer coded quantum particle swarm optimization algorithm which takes the minimum network loss and node voltage stability as the objective function. First, a day of load forecasting curve is divided into several time intervals based on its monotonicity and amplitude changes. Then integer quantum particle swarm optimization algorithm is adopted to realize dynamic distribution network reconfiguration. In order to find the best number of refactoring, the optimal time refactoring function is proposed based on adjacent time refactoring changes value network loss. In consideration of the power distribution network loss minimum and node voltage fluctuations largest and smallest for the target, we are looking for best reconstruction of the distribution network structure on the basis of IEEE33 distribution system, the reconstructed results verify the validity of the proposed method.