This paper proposes a fault diagnosis method for rolling bearings under variable speed conditions， based on the Adaptive Window Rotation Optimization Short-Time Fourier Transform （AWROSTFT）. This method addresses the issue of low energy concentration caused by the fixed window effect in Short-Time Fourier Transform （STFT）. Variational Mode Decomposition （VMD） is used to reduce the noise of the original vibration signal， and Particle Swarm Optimization （PSO） is employed to solve the complex problem of VMD parameter selection. A series of rotation operators are adaptively matched to the horizontal window in STFT using the tangent idea， aligning the rotation direction of the window with the instantaneous frequency modulation to im? prove the energy concentration of time-frequency representation. The instantaneous frequency， extracted by the spectral peak detec? tion method， is divided by the frequency transformation curve. The result is matched with the fault characteristic coefficient of the bearing to achieve fault diagnosis of the rolling bearing under variable speed conditions. The results of simulation and experimental signals show that the proposed method effectively combines the advantages of PSO-VMD and AWROSTFT. Through the adap? tive rotation window with the idea of tangency， the angle between the signal and the window function is globally reduced to zero， improving energy concentration， sharpening the time-frequency ridge line， and enabling fault diagnosis of rolling bearings under variable speed conditions.