The influences of network reconfiguration and distributed generations (DGs) on distribution system state estimation (DSSE) and measurement placement should not be ignored in active distribution systems. In this paper, considering network reconfiguration and the output uncertainties of DGs, a robust measurement placement method for active distribution systems is proposed based on measurement saturation analysis and heuristic algorithm. First, the saturation number determined by measurement saturation characteristic is chosen as the measurement number. Then, the impacts of different network topologies are represented by various weights in the robust measurement placement model, which are computed by Markov chain and analytic hierarchy process. In addition, Gaussian mixture model is applied to approximate the power fluctuations of DGs. Uncertainties caused by measurements are considered by Monte Carlo simulations. The accuracy of DSSE in different network topologies can be guaranteed by the proposed robust measurement placement method. Simulation results based on the IEEE 33-bus and 119-bus distribution systems demonstrate the effectiveness of the proposed method.