The rectifier diodes in the rotating rectifier of aviation three-stage generator are exposed to high-speed rotation and high-temperature working environments for a long time, enduring extremely strong centrifugal and thermal stresses, which are prone to aging and failure, which will directly affect the operational safety of the generator. Therefore, this paper focuses on the life prediction research of rectifier diodes. The forward voltage drop of the rectifier diodes is selected as its aging characteristic parameter. Based on a large amount of measured data on the diode's forward voltage drop obtained from the self-developed diodes accelerated aging test platform, a genetic algorithm optimized back propagation neural network (GA-BP) algorithm is used to predict the lifespan of the rectifier diodes. The research results indicate that this prediction method can predict future aging trends based on the historical data of rectifier diodes, and detect the aging risk of rectifier diodes in advance successfully.