Abstract: The Three Gorges ship lift employs a vertical ship lift model using gears and racks, with the racks serving as the transmission component for the driving equipment of the ship lift during lifting operations. The safety and reliability of the equipment are crucial for the overall safety of the ship lift operation. Therefore, it is essential to establish an intelligent online state monitoring system for this purpose. This study constructs a small-scale test bench to simulate various defect conditions and utilizes the frequency spectrum slice analysis method based on synchronous compression wavelet transform and the fault diagnosis method of gated cyclic neural network to detect and identify issues such as poor lubrication, tooth surface pitting, and tooth root cracks in the racks. The experiments demonstrate the effectiveness of this diagnostic approach. Building upon these findings, an online condition monitoring scheme for the pinions and racks used in the Three Gorges ship lift is proposed, aiming to provide technical support for the intelligent management and maintenance of the lift.