In order to smooth out the volatility of distributed power sources after grid connection and optimize the operation and management of the distribution network after grid connection. A two-stage coordinated and optimized operation strategy of active and reactive power was proposed for a new type of power system under source-grid-load-storage interaction. Along with research on distributed power source siting and capacity determination. The first stage adjusts the ratio of on-load regulator transformer and shunt capacitor bank, with network loss and voltage deviation were taken as the optimization targets. The second stage adjusts the static reactive power compensator and photovoltaic inverter, the target is the fluctuation of active power at the photovoltaic grid-connected point and the minimum voltage deviation at the grid-connected point. The two stages carry out information transfer through iterative coordination and feedback methods, combined with uniform network theory and improved multi-objective cyclic small habitat particle swarm optimization algorithm, and verified by simulation on the IEEE33 node system, which shows that the voltage quality lifting and network loss reduction as well as suppression of PV fluctuation have obvious effects. Finally, the study investigates distributed power source siting and capacity determination, further illustrating the enhanced algorithm’s versatility.