Nonlinear parameters identification of mean value engine models based on neural network

In this paper, different Neural Networks were used to identify the intake-branches nonlinear parameters of mean value engine models which meant effective flow area of throttle and volumetric efficiency of the cylinder. In this way, the air flow rate could be pre-estimated more accurately. Compared with the methods used before, the accuracy of the model was improved greatly, therefore, the control of EFI engine based on the model could be more widely used.