Research on the Prediction of Gasoline Engine Air Intake Flow Based on the BP Neural Network

In order to improving the real-time of air-fuel ratio control system and accuracy of inlet flow measurement for gas line engine, this paper proposed chaotic time series BP neural network prediction model for engine´s inlet flow. By using phase space reconstruction technique of inlet flow time series of refactoring, recovery system of the original chaotic. Using BP network to training and projections reconstructed data, to improve the inlet flow measurement accuracy, and thus improve the gasoline engine air-fuel ratio control system of the real-time and accuracy. Experimental simulation results show that the prediction model of Chaotic sequence BP neural network has higher prediction accuracy, and provides a new approach for accurate and timely test gasoline engine inlet flow.