Application of Neural Networks for Prediction and Optimization of Emissions and Performance in a Hydrogen Fuelled Direct Injection Engine Equipped With In Cylinder Water Injection

In this work Artificial Neural Networks (ANN) technique has been used to predict NOx emissions and Indicated Thermal Efficiency (ITE) for a direct injection Hydrogen engine, which is equipped with water direct injection system for NOx control. ANN has been used as a mathematical tool that learns from the experimental data obtained under different operating conditions. Feed forward multilayer perception network is used for nonlinear mapping between the input and output parameters. Different backpropagation training algorithms, activation functions and several rules are used to assess the percentage error between the target and the predicted values. As good correlations between measured and predicted NOx emissions and engine ITE are obtained, a step further using the ANN as an optimization tool has been performed. It has been found that, higher values of ITE can be achieved at equivalence ratio range from 0.4 to 0.9 where outside this range there is no improvement in the ITE even with direct water injection. On the contrary, the NOx reduction is achieved with direct water injection at all Hydrogen/air equivalence ratios.