Feasibility Study on the Use of Artificial Neural Networks to Model Catalytic Oxidation in a Metallic Foam Reactor

Abstract

This study investigates the feasibility of using artificial neural networks (ANNs) to predict catalytic oxidation in diesel after-treatment systems and compares their performance to that of physics-based models. Existing physics models are revisited to generate baseline data for binary reactions of major species (CO, C 3 H 6 , and NO) measured in a lab-scale microreactor comprising a metallic foam catalytic substrate. The physics model performs well to predict the measured light-off curves, which are the species conversions with ramping temperature, and the R 2 value is above 0.84 across a wide range of operating conditions. However, the model cannot perfectly capture the retarding trends observed in the CO and C 3 H 6 conversion curves after light-off. In contrast, the ANN model is capable of accurately predicting the light-off curves for operating conditions seen during the training process. This might be practically useful but is inherently limited by the availability of experimental data for training. To compensate for the drawbacks of both approaches, this study suggests a hybrid model in which a pretrained ANN is used to calculate reaction rates in the physics models. Despite the more complex data generation process for training ANNs, the hybrid model captures the light-off curves including the retarding trend and is less sensitive to the range of test conditions without renormalization as compared to the pure ANN model. This study investigates the feasibility of ANNs by comparing the pros and cons among the physics models, pure ANN, and hybrid models and suggests a step toward the most appropriate uses of ANNs in modeling exhaust after-treatment in practical applications.

Review of Machine Learning for Hydrodynamics, Transport, and Reactions in Multiphase Flows and Reactors

“…In some specific systems, it has been designed to compensate for the weakness of both the pure ML model and the physical model. 21 (3) The domain knowledge-informed ML model: This is an emerging learning method that can integrate the prior physics, mathematical laws, chemical mechanisms, or boundary conditions as constraints into the architecture of ML algorithms. 18 The role of such constraints is to teach the pure ML models about the prior knowledge, which can not only greatly improve its approximation ability but also boost the interpretability that the pure ML (especially NN) does not have.…”

“…The hybrid model is also applicable for predicting the input–output relationship problems mentioned in (1) and has a potential to achieve better model performance than the pure DDM. In some specific systems, it has been designed to compensate for the weakness of both the pure ML model and the physical model …”