Abstract: The Neural Network is one of the most important artifical intelligence technology at present． It simulates the structure of human brain and its function of dealing with or memorizing the information． The Neural Network is adept at learning from the input and output data and draw out an useful information．The predicting of the performance of engine is to predict the engine performance using the structure and operation parameters of the engine．So we can use the learning peculiarity of Neural Network to show how the structure and operation parameters of the engine affect the performance of engine‚and describe them with a Network model．The authors of this article discusse how to choose the suitable network and uses it to construct the model of engine instead of physical model‚then give an example of an engine model of emissions‚and programs on MATLAB.

Key words: generalized regression neural network, engine, forecast model