Submersible motors are widely used in industry, agriculture, and life. Despite their safety and efficiency, submersible motors suffer from a harsh working environment. Overheating and overload may cause seal damage to make sediment flood in, which will lead to bearing faults. Meanwhile, due to the particularity of the working environment of submersible motors, the traditional method of detecting bearing faults through vibration signals is challenging to implement. In this paper, a numerical model based on the eccentricity analysis method is proposed to simulate the bearing faults process of submersible induction motors so that electrical signals, including air gap flux density and exciting force, could be obtained to diagnose the bearing faults. The proposed approach is capable of approximately simulating the motion orbit of the rotor origin when a single point fault occurs. Then, the 2D Fourier decomposition method is used to analyze the influence of rotor deflection on air gap flux density and exciting electromagnetic force from order and frequency. Finally, the proposed model is tested and proved efficient by analyzing the emerging fault harmonic components. © 2022 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.