This paper presents a new uncertainty analysis method of static voltage stability for power systems with wind power generators

based on the evidence theory and bilevel programming. Firstly

the uncertainty of wind speed is described by the concept of basic probability assignment(BPA)in the evidence theory

and the uncertainty analysis model of the static voltage stability is developed. Then

a plurality of hypercube domains are obtained by constructing the joint basic probability assignment of the wind speed

and the original uncertainty analysis problem of static voltage stability is transformed into the problem of farthest and closest voltage stability critical point in these hypercube domains. Finally

bilevel programming method is applied to calculate the farthest and closest critical point problem

and the maximum and minimum load margin in these hypercube domains are obtained

then the belief cumulative probability distribution(BCPD)and plausibility cumulative probability distribution(PCPD)of load margin are synthesized. The numerical results of IEEE 118-bus and IEEE 300-bus systems indicate that

compared with traditional probabilistic methods

the proposed model and method can describe the objective and subjective uncertainty information of wind speed

and the minimum and maximum probability of voltage instability can be determined through the BCPD and PCPD of the load margin. It is in line with the thinking habits of power dispatching personnel.