Application research on the model identification with neural networks

A method to identify the efficiency factor of the automobile motor is described based on the principle of neural networks. It is supposed that this factor depends on the angle velocity of the motor axis and the pressure in the collector. The data measured is divided into two groups, one is used for the identification, the other for the validation determining the capacity predictive of this model. By comparing the two models built up with a common method and neural networks, it is possible to conclude that the latter is effective and the result is good.