In order to effectively utilize geographic information technology to support the information construction of complex large power grid, the impact of blackouts on power system operation and daily life, an optimal outage model based on deep artificial neural network and GIS data is proposed. Combining the particularity of power system operation, the optimal parameter setting is united with incremental feedback to optimize constrained Boltzmann algorithm. The performance of the algorithm is analyzed by simulation, and simulation results show that the optimal power outage model using deep neural network can improve the efficiency and accuracy of computation. This work is supported by National High-tech R & D Program of China (863 Program) (No. 2011AA05A116) and Science and Technology Project of State Grid Corporation of China (No. KJ00-01-08-02).