Heat transfer coefficients of flow condensation under microgravity are the basis for design of aerospace heat exchanger， but their experimental data are scarce； thus it is necessary to establish an accurate prediction model. In the present study， a prediction model of flow condensation heat transfer under microgravity is established based on artificial neural network. Two kinds of neural networks， i.e. back propagation （BP） and radial basis function （RBF） are adopted； the hydraulic diameter， saturation temperature， mass flux， vapor quality and parameter related to fluid thermophysical properties are used as network inputs， while the condensation heat transfer coefficient is taken as network output. The results show that for BP neural network model， the root mean square error （RMSE） and mean absolute percent error （MAPE） are 237 and 4.32%， respectively； while for RBF neural network model， RMSE and MAPE are 165 and 2.35%， respectively. Compared with BP neural network， RBF neural network has the higher accuracy. The flow condensation heat transfer cofficients predicted by RBF neural network model agree with 94% of experimental data and numerical results within the deviation of ±10%.