A thrust-vectoring nozzle modeling method is developed by means of the Finite Volume Method (FVM) and Back-Propagations neural networks.The neural networks learning samples are the results of the FVM computation.The obtained model describes the functions of the nozzle thrust

thrust coefficient and vectoring angle to the inlet

exit and structural parameters of the nozzle.The model is of great significance for reseaching vectoring nozzle control law and Integrated Flight/Propulsion control.The investigation shows that the deviations among the output of the model

the results of FVM computation

and the test data are quite small.