Abstract: To accurately grasp the transient stability of voltage in grid connected power systems containing renewable energy, this paper proposes an improved evaluation method based on the support vector machine for ball loss. Firstly, based on the theory of system characteristics and projection energy function indicators, establish the original feature set of power system transient stability margin; Subsequently, using the maximum correlation minimum redundancy feature selection method, a subset of features that measure the transient stability margin of the grid connected power system voltage is selected to characterize the impact of renewable energy grid connection on the transient stability margin of the power system voltage; Thirdly, the quantile method is used to divide the system into stable and unstable classes, and the classification problem of voltage transient stability margin is transformed into the problem of finding the optimal quantile distance in PEF PinSVM, reducing the impact of renewable energy grid connection volatility and uncertainty on the evaluation of system voltage transient stability margin; Finally, using the IEEE39 node system as an example simulation, it is found that the accuracy of voltage transient stability margin assessment in grid connected power systems containing renewable energy is imporved with the increase of renewable energy penetration rate.