The optimal PMU placement (OPP) as a prerequisite for real-time state estimation of active distribution networks and other power system networks is important for situational awareness of power systems containing large amounts of dynamic distributed renewable energy. However, existing OPP solutions based on heuristic intelligence algorithms have problems such as high algorithmic complexity, the tendency to fall into local optima and uncertainty in the output. In this paper, PMU monitoring equipment deployment method based on swarm intelligence algorithm considering special nodes with two-stage optimization search is proposed for the OPP problem. The principles of PMU configuration for special nodes in the power system network are first summarized and priority is given to these special nodes in the algorithm, and then the optimal number of deployment points and monitoring node configuration scheme is obtained through a two-stage optimization search. The performance of the algorithm is verified based on IEEE standard algorithms. Compared with existing integer planning methods, this algorithm achieves optimal configuration and higher redundancy of measurements on the IEEE-30 node system, thus reducing the cost of deployment of condition-aware monitoring devices, improving the accuracy of condition-awareness, and providing a basis for situational awareness of distributed energy sources with uncertainty into the active distribution network.