Hydro-turbine governing system is a time-varying complex system with strong non-lineafity,and its dynamic char-acteristics are jointly affected by hydraulic,mechanical,electrical,and other factors.Aiming at the stability of the hydro-turbine governing system,this paper first builds a dynamic model of the hydro-turbine governing system through mecha-nism modeling,and introduces the transfer coefficient characteristics under different load conditions to obtain the stability category of the system.BP neural network is used to perform the machine study and the predictive analysis of the stability of the system under different working conditions is carried out by using the additional momentum method to optimize the algorithm.The test set results show that the method can accurately distinguish the stability category of the hydro-turbine governing system (HTGS),and the research results can provide a theoretical reference for the operation and management of smart hydropower stations in the future.