With the development of large-scale renewable energy power grids, the risk of cascading failures is becoming increasingly higher. Appropriate intentional islanding can prevent the spread of faults. To solve the problems of topological connectivity of power systems and high computational complexity in the process of intentional islanding sections, this paper proposes an intentional islanding section search method with a topological connectivity constraint genetic algorithm. The method first constructs a mathematical model for intentional islanding section searching, and then proposes a system topology connectivity constraint based on the knowledge of graph theory. This is added to the genetic algorithm. At the same time, a system topology simplification and preprocessing method is proposed. Finally, the simplified system topology and data is input into the genetic algorithm with topological connectivity constraint to solve the intentional islanding sections. The calculation example on the IEEE 118 node system shows that the proposed algorithm can ensure that the solved intentional islanding sections has topological connectivity and basically achieves active power balance, and the system topology simplification and preprocessing methods can effectively improve the efficiency of the algorithm. This work is supported by the Science and Technology Project of the Headquarters of State Grid Corporation of China (No. 5211DS21N013).