The one-dimensional setting of the classical stochastic resonance problem was extended to the two-dimensional potential field, the mechanism and application of the stochastic resonance of a two-dimensional bistable potential system under the synergy of noise and weak external driving force were studied.Based on the adiabatic approximation theory, the analytical formula of the output signal-to-noise ratio was derived, and it is found that the significant stochastic resonance phenomena occur in the system.The output signal-to-noise ratio can be further improved by increasing the amplitude of the driving force, the coupling coefficient, the correlation coefficient, or reducing the driving frequency.The fourth-order Runge-Kutta method and the genetic algorithm were used for numerical simulation.It is noted that the theoretical curves of the output signal-to-noise ratio basically coincide with the actual numerical curves.The weak signal detection experiment results show that the system can effectively detect weak periodic signals.The system was applied to the practical application of industrial bearing fault diagnosis.The results show that the system can effectively diagnose the faults of the bearing inner and outer rings.By comparing the peak value and the recognition of the peak value, it is proved that the fault diagnosis effect of the system is better than that of the novel one-dimensional unsaturated bistable system.