Published in Neural Networks: Overview of Current Theories and Applications, Smart Innovation, Systems and Technologies, Springer, pages 247-260 (DOI 10.1007/978-981-95-4072-3_21).
The paper compares machine learning models (linear regression, SVM, decision trees, XGBoost, MLP, LSTM, GRU and KNN) for predicting carbon monoxide and nitrogen oxides emissions from gas turbines as part of a predictive emission monitoring system. It gives a side-by-side view of which methods estimate these emissions most accurately.
Authors: Kamyar Zeinalipour, Laure Barriere, David Ghelardi, Marco Gori.