Artificial intelligence video surveillance technology has broad application prospects in substation safety monitoring and management. In order to improve the utilization efficiency of the camera terminal in the video surveillance system of smart substation, the optimization layout model and algorithm of the smart camera terminal are studied. Firstly, based on the idea of rasterization, the plan of the substation is divided into regions. And the importance matrix of monitoring area is proposed to describe the importance of regions and equipment. Then, a definition discrete model of camera monitoring range is established. Based on this, the monitoring multiplier function is introduced to establish the layout optimization model for substation surveillance cameras terminals. This model takes the minimum average monitoring distance as the optimization goal, taking into account the constraints of camera selection and site selection and acquisition cost. And then the genetic algorithm is used to solve the model to get the optimal layout of the cameras. Finally, the validity and practicability of the model in reducing the number of monitoring blind spots are verified by simulation example, which has high engineering application value. This work is supported by Science and Technology Project of Guangdong Power Grid Ltd. (No. 031800KK52160013) and Natural Science Foundation of Guangdong Province (No. 2017A030313304).