Precise measurement of flush air data for advanced flying-wing aircraft without external atmospheric probes is a vital task in aircraft design. This work presents a novel artificial neural network (ANN) model for Flush Air Data Sensing (FADS) System based on high-fidelity computational fluid dynamics (CFD) simulations. In the NN model, four pressures measured on different locations of the fuselage are taken as input and free stream pressure, the Mach number and the angle of attach (AOA) are taken as output. It is shown that a wide range of Mach numbers of \(Ma = 0.2{-}0.8\) and angles of attach \(AoA = 0{-}15^{ \circ }\) are accurately obtained by the proposed ANN model with four inputs. Particularly, it is also found that the Mach number sensed based on pressure distribution can be used to calculate the total pressure, instead of the conventional method of directly measuring total pressure through an airspeed tube. Finally, applications and validations of the proposed ANN model are carried out by numerical simulations to show its potential applications in the flying-wing aircraft.