With the increase of the proportion for renewable energy sources access to islanded microgrid

the power flow calculation of the islanded microgrid under the traditional deterministic conditions is challenged. Based on the two droop control strategies of active power-frequency/ reactive power-voltage(P-f/Q-U)and active power-voltage/reactive power-frequency(P-U/Q-f)

a probabilistic power flow model considering the uncertainty of renewable energy for islanded AC microgrid under comprehensive control is established. Meanwhile

a modified three-step Levenberg-Marquardt(MTLM)algorithm is proposed to solve the power flow equation. A quasi-Monte Carlo simulation based on the Sobol sequence is used to obtain the samples of stochastic renewable energy and loads

then the probabilistic load flow calculation of the 38-node islanded microgrid is carried out. The fast convergence and robustness of the MTLM algorithm are verified by comprising with other methods. Moreover

the probability distribution functions of frequency and voltage under different droop control strategies are studied

and the influence of stochastic renewable energy access to islanded microgrid is analyzed.