Abstract:In order to reduce the experimental cost of establishing engine performance prediction model, machine learning algorithms with powerful non-linear problem analysis ability were used to predict the performance of GDI turbocharged engine, such as General Regression Neural Network (GRNN) and Support Vector Regression (SVR). At first, the Taguchi orthogonal experiment method and the Latin Hypercube Sampling (LHS) were used to determine the operation points of the training data and the test data. And then based on the same training data training model containing only 25 samples, the prediction performance of GRNN and SVR was tested and compared using 100 sets of identical test data. The comparative analysis shows that GRNN has the risk of converging to the local minimum when the experimental data is limited, and SVR can find the optimal global solution with good prediction accuracy and generalization ability. So SVR is very suitable for GDI Turbocharged engine performance predictions and would significantly reduce experimental costs