Search

Developing an accurate combustion model for a Sequential Turbocharging diesel engine is an efficient way to reduce the time consumption and the expensive experiment costs in the process of determining the control strategy of a Sequential Turbocharging system. In this paper, 142 operations of a marine diesel engine with two Sequential Turbocharging systems are used for this research. A novel method is proposed to calculate the initial values and limit bounds of the Wiebe parameters, and the use of this method could effectively avoid the occurrence of impossible solutions or multiple solutions caused by only using nonlinear least square fitting. The calculated Wiebe parameters could achieve a high level of accuracy that the mean value of the determination coefficient between the simulated and experimental heat release rate profiles is 0.98 for all the operations, and they are accurate enough for developing the combustion model. The partial least square regression is first imposed to relate the Wiebe parameters to the operating conditions. For the ability to solve the multiple correlations among the operating conditions, the prediction performance of the combustion model developed by partial least square regression is much higher than that by least square regression. The mean absolute percentage errors of the brake specific fuel consumption and maximum in-cylinder pressure simulated by the model are 0.284% and 1.39%, respectively. The model could accurately predict the engine performance and is sufficient to make the reasonable control strategies for the different Sequential Turbocharging systems.

In this work, two kinds of partial least squares modelling methods are applied to predict a compressor map: one uses a power function polynomial as the basis function (PLSO), and the other uses a trigonometric function polynomial (PLSN). To demonstrate the potential capabilities of PLSO and PLSN for a typical interpolated prediction and an extrapolated prediction, they are compared with two other classical data-driven modelling methods, namely the look-up table and artificial neural network (ANN). PLSO and PLSN are also compared with each other. The results show that PLSO and PLSN have a better prediction performance than the look-up table and the ANN, especially for the extrapolated prediction. The computational time is also decreased sharply. Compared with PLSO, PLSN is characterized by a higher prediction accuracy and shorter computational time than PLSO. It is expected that PLSN could save computational time and also improve the accuracy of a thermodynamic model of a diesel engine.