The static voltage stability assessment method of a power system based on data drive usually has the problem of an unbalanced sample category in the initial data，which makes the performance of the data-driven assessment model greatly affected.Therefore，a static voltage stability assessment method of a power system is proposed based on the majority weighted minority oversampling technique （MWMOTE ） and sparrow search algorithm-kernel extreme learning machine （SSA-KELM ）.First，MWMOTE is used to solve the problem of an unbalanced sample category and increase sample diversity.Then，the KELM model parameters are optimized by using SSA，and the static voltage stability assessment model of the power system based on SSA-KELM is constructed.Finally，the validation is carried out on a 10-machine 39-bus system of New England，and the test results show that the proposed method can effectively deal with the problem of unbalanced sample categories with good assessment accuracy and generalization ability.