Abstract: The active distribution network contains various distributed generation, electric vehicles, energy storage equipment and flexible loads, which pose challenges to the operation and maintenance plan of the grid enterprise due to its uncertainty in the domain of time and space. From the point of service restoration after a blackout, this paper considers the uncertain factors of wind and photovoltaic in active distribution networks and uses electric vehicles as emergency dispatching resources. A three-layer robust islanding recovery mathematical model in the form of min-max-min is established which ensures the minimum load shedding in the worst case. In terms of solving method, the three-layer robust optimization model is con v e r t e d to a re l a xe d ma in pr o b l e m an d a ma x-m i n sub-problem using the Column ＆ Constraint Generation（CCG）method. By adding feasible cuts and optimal cuts iteratively into the relaxed main problem, the optimal solution of the original problem is obtained. Considering the non-convexity of t h e D i s t f l o w p o w e r f l o w e q u a t i o n s i n t h e m a x-m i n sub-problem, the sub-problem is converted into a MISOCP pr o gr a m m i ng mo de l by us i n g re l a x a t io n te c hn i q u e s.Meanwhile, the duality theory is utilized to transform the max-min sub-problem into a single-layer max problem. Finally,the effectiveness of the proposed model and method is verified by the simulations of PG＆E 69 distribution network system.