An interturn short-circuit fault diagnosis strategy for the interior permanent-magnet synchronous motor (IPMSM) of an electric vehicle (EV) was presented. Online and offline detection methods were integrated into the diagnostic strategy to optimize diagnostic performance based on work cycle of an EV. An iterative form was used to reduce the computational load in online detection. It can be embedded into the control program of an inverter using no additional hardware. To improve the diagnostic performance, high-frequency voltage excitation was employed to amplify the fault information in offline detection. Good performance for a variety of work cycles was achieved by the integration of the two detection methods. Experiments were presented to verify the proposed diagnosis strategy.