Prediction of NOx emissions by using the cylinderpressure curves of an internal combustion engine

Hello everybody,

I am looking for the right tool to predict NOx emissions based on the cylinder pressure curve.

For example, 50 pressure values ​​depending on the crank angle are recorded for each cycle. This time-series respectively curve is the input. For this pressure curve there is the associated NOx value as the output parameter. Further inputs should not be considered. The aim is to predict the NOx emissions for each pressure curve.

Question:

In which tool can I use curves as input? Taking into account the time dependency of each cycle.

Which math models and which Matlab libraries are the right ones to take into account the dependencies over time (radial base models, neural networks, or maybe DeepLearningToolbox)

Thank you very much!

Answer 1

[Emmanouil Tzorakoleftherakis]

This sounds like a supervised learning problem with time dependencies in which case I would recommend working with LSTMs and Deep Learning Toolbox. Here are a couple of links that seem relevant to what you were mentioning:

https://www.mathworks.com/help/deeplearning/ug/long-short-term-memory-networks.html#mw_ed693e72-a264-4b57-877c-c9b7890b79e1

https://www.mathworks.com/help/deeplearning/ug/sequence-to-sequence-regression-using-deep-learning.html

https://www.mathworks.com/help/deeplearning/ug/time-series-forecasting-using-deep-learning.html