Abstract: In order to effectively assess drought risk, the uncertainty divergence term in the quintuple correlation number is decomposed into partial similarity divergence and partial dissimilarity divergence terms, resulting in an evolutionary quadruple subtraction set for potential, and the divergence coefficients are subjected to triangular fuzzy number stochastic simulation. A drought risk assessment model based on evolutionary quadruple subtraction set for fuzzy number stochastic simulation is established and applied to the drought risk assessment in Huainan City. The results show that from 2008 to 2017, the overall drought risk level in Huainan City ranged from level 2 to level 3, indicating a moderately low risk state, but there was a slight upward trend in risk level after 2014. In the subsystems of drought risk assessment, the risk levels of the hazard subsystem, exposure subsystem, and vulnerability subsystem gradually decreased, showing a positive development trend, while the resilience subsystem level gradually increased. Factors such as per unit area water resources, irrigation index, replanting index, ratio of paddy field area, and forest coverage rate were identified as the main obstacles. The drought risk assessment method based on evolutionary quadruple subtraction set for fuzzy number stochastic simulation includes the calculation results of quintuple subtraction set for potential, and its calculation results form a grade confidence interval, which can comprehensively reflect the multi-factor influences and uncertainties of drought risk changes.