The NOx and SOx output of a combined heat and power plant is studied with the aim of replacing the physical sensor array with a mathematical formula that can compute the emissions rather than measure them. The model is determined using machine learning from historical measurements and uses neural networks. As the model can be cheaply deployed, will not drift, and will not malfunction or fail, the model has significant added value over a physical sensor. We find that the accuracy of the model is comparable to the accuracy of the measurement and is thus suitable for a full replacement.
Number of Pages
California Energy Commission (2015): Guidelines for Certification of Combined Heat and Power Systems Pursuant to the Waste Heat and Carbon Emissions Reduction Act. CEC-200-2015-001-CMF.
Goodfellow, I., Bengio, Y., Courville, A. (2016): Deep Learning. MIT Press.
Hastie, T., Tibshirani, R., Friedman, J. (2016): The Elements of Statistical Learning: Data Mining, Inference, and Prediction. 2nd Edition. Springer.
Yu, Y., Si, X., Hu, C., Zhang, J. (2019): A Review of Recurrent Neural Networks: LSTM Cells and Network Architectures. Neural Computation 31, 1235-1270.
Parmar, A., Katariya, R., Patel, V. (2018): A Review on Random Forest: An Ensemble Classifier. Proceedings of ICICI 2018: International Conference on Intelligent Data Communication Technologies and Internet of Things (ICICI). Springer.
Bangert, P. (2012): Optimization for Industrial Problems. Springer.
Guyon, I., Elisseeff, A. (2003): An Introduction to Variable and Feature Selection. Journal of Machine Learning Research 3, 1157-1182.
Bishop, C.M. (2006): Pattern Recognition and Machine Learning. Springer.
Tsai, C.L., Cai, Z., Wu, X. (1998): The Examination of Residual Plots. Statistica Sinica 8, 445-465.
Looking for more?
Some of the OnePetro partner societies have developed subject- specific wikis that may help.