To improve distribution network reconfiguration with Distributed Generation (DG), a dynamic distribution network reconfiguration model considering time-varying characteristics of DG output and load demand is established. First, the K-means++ clustering algorithm is used to divide the daily load period. Then, the system loss and voltage deviation are taken as the objective functions, and the improved grey wolf optimization algorithm is used to optimize the calculation. To tackle uneven initial population distribution, lack of global communication and ‘easy to fall into local optima’ in traditional grey wolf optimization algorithms, when generating the initial population, it introduces tent mapping to enhance the uniformity of the initial population. A cooperative competition mechanism is introduced to improve the utilization rate of effective information between individuals. An adaptive inertia weight is introduced when the grey wolf population position is updated to meet the optimization requirements of different periods. Finally, the feasibility and superiority of the proposed algorithm are verified by a numerical example. This work is supported by the National Key Research and Development Program of China (No. 2017YFB0902800) and the Science and Technology Project of State Grid Corporation of China (No. 52094017003D).