Comparison of CNN and LSTM for modeling virtual sensors in an engine 2020-01-0735
Automotive industry makes extensive use of virtual models to increase the efficiency during the development stage. The complexity of such virtual models increases as the complexity of the process that they describe, and for this reason new methods for their development are constantly evaluated. Among many others, data-driven techniques and machine learning offer promising results, creating deep neural networks that map input-output relations. This works aims at evaluating the performance of different neural network architectures for the estimation of engine status and gas emissions. More specifically, Convolutional Neural Network (CNN) and Long-Short Term Memory (LSTM) will be evaluated in terms of performance, using different techniques to increase the model generalization. During the learning stage data from different engine cycles are fed to the neural network.
To evaluate model generalization the network is then tested over new, previously unseen, engine cycles. Results show that our model over-performs a state of the art models, the best performance was found from the LSTM model with 2.40%, 2.80% and 18.19% error for flow fuel, NOx and soot sensor respectively.