NOx Prediction by Cylinder Pressure Based on RBF Neural Network in Diesel Engine

To meet electronic control technology demand based on cylinder pressure feedback in diesel engine, prediction of cylinder pressure feedback variable based on Radial Basis Function (RBF) neural networks is made. Briefly analyzed disadvantage of curve fitting method by multi-parameter input mapping single output, radial basis function neural networks is introduced, faster algorithm of Orthogonal Least Squares (OLS) is adopted to calculate networks. Prediction model of cylinder pressure feedback variable based on radial basis function neural networks is present by using Nitric Oxide (NOx) as example, training time and prediction precision is analyzed, comparing with BP neural networks, verification of prediction result by RBF neural networks is made. Test result is shown that prediction model of cylinder pressure feedback variable based on radial basis function neural networks can meet the requirement of diesel engine.