A New Approach to System Level Soot Modeling

A procedure has been developed to build system level predictive models that incorporate physical laws as well as information derived from experimental data. In particular a soot model was developed, trained and tested using experimental data. It was seen that the model could fit available experimental data given sufficient training time. Future accuracy on data points not encountered during training was estimated and seen to be good. The approach relies on the physical phenomena predicted by an existing system level phenomenological soot model coupled with ‘weights’ which use experimental data to adjust the predicted physical sub-model parameters to fit the data. This approach has developed from attempts at incorporating physical phenomena into neural networks for predicting emissions. Model training uses neural network training concepts. Similarity in the final weight vectors between different models trained on different sets of data suggests that the weights are adjusting the predicted (by the phenomenological model) physical (sub)quantities in a general manner to capture unaccounted for phenomena particular to the engine. The model can be a useful tool for engine/aftertreatment control, design and optimization, since it delivers reliable results and is computationally inexpensive. For example, the model developed in the current work has been used to feed into a diesel particulate filter (DPF) model as part of a larger integrated engine-aftertreatment system level model. Finally, a procedure was developed to estimate the most useful future experimental data points for improving model performance.