This article delves into the application of vibration environmental prediction methods based on the Power Spectral Density Transfer Matrix (PSDT) in APU subsystems. Initially, the estimation methods of PSDT are optimized, and the definition and calculation method of vibration environmental migration coefficients are further introduced, thereby deriving a vibration environmental prediction model for the subsystem. To validate the effectiveness of this improved PSDT estimation method, a simulation experiment involving a lateral vibration beam with 5 degrees of freedom was designed, successfully confirming the accuracy of this approach. Furthermore, vibration test design was conducted using a typical APU system as an example. Through the analysis of experimental data, the reliability of the improved PSDT estimation method was reconfirmed. Ultimately, the vibration migration coefficients and vibration environment of the APU subsystem under three different directions and four flight phases under windmill conditions were successfully obtained. This method utilizes acceleration response data from systems under actual operating conditions to predict vibration environments for subsystems, which is not only straightforward to operate but also has significant practical application value for vibration environmental prediction and design input of aircraft onboard equipment subsystems.