The extended Monte Carlo method is capable of translating the error in the probabilistic load flow into controllable parameters and retaining the power flow results already obtained with the sample size increased. The conventional method based on the Latin hypercube sampling is more efficient than simple random sampling

but still have two main drawbacks: firstly

the Latin hypercube sampling method is of low accuracy for its incapability of generating sampling sequences of low discrepancy

which constitutes the bottleneck of convergence; secondly

there is no suitable convergence criterion that can be adopted in the non-normal distribution scenarios. In order to overcome these two drawbacks

a probabilistic load flow method based on Sobol sequence is proposed. Furthermore

a convergence criterion based on non-parametric density estimator is employed. The simulation results on IEEE 30-bus system and IEEE 118 bus system demonstrate the validity of the proposed method. In contrast to the method based on Latin hypercube sampling

the proposed method is of high efficiency

accuracy and speed. And the adoption of the convergence criterion is more direct and flexible

and better able to guide the convergence in extended probabilistic load flow. This work is supported by State Key Program of National Natural Science Foundation of China (No. 51337005) and National Basic Research Program of China (973 Program) (No. 2014CB23903).