OVERVIEW OF EXISTING SOLUTIONS DESIGN PREDICTIVE EMISSION MONITORING SYSTEMS PEMS
DOI:
https://doi.org/10.32339/0135-5910-2026-8-49-61Keywords:
predictive emission monitoring system, nitrogen oxide, sulfur dioxide, carbon monoxide, machine learning, artificial neural networksAbstract
In response to tightening environmental legislation and economic constraints, Predictive Emission Monitoring Systems (PEMS) have emerged as a key technology for estimating pollutant emissions, particularly nitrogen oxides, sulfur dioxide, and carbon monoxide. Unlike instrumental Continuous Emission Monitoring Systems (CEMS), PEMS rely on mathematical models that predict emission levels based on process parameters. This review presents a comparative analysis of representative studies on emission modeling from power generation units using various machine learning techniques, ranging from artificial neural networks (ANNs) and support vector regression (SVR) to linear regression and gradient boosting (XGBoost). The aim is to identify the evolution of approaches and compare them in terms of accuracy, speed, stability, interpretability, and suitability for industrial deployment. The analysis concludes that modern ensemble methods based on decision trees, particularly XGBoost, offer the best balance of characteristics, providing high accuracy, compliance with regulatory requirements, good interpretability, and robustness to sensor failures. Nevertheless, for simple process configurations, linear models remain a robust and practical solution. The review also outlines key directions for future research, including improving predictions under extreme operating conditions, integrating dynamic and adaptive components, and enhancing model generalization capabilities.
References
Hung W. S. Y. An experimentally verified NOx emission model for gas turbine combustors. ASME International Gas Turbine Conference and Products Show, 1975, vol. 1B, 11 p. DOI: 10.1115/75-GT-71.
Chien T. W., Chu H., Hsu W. C. A performance study of PEMS applied to the Hsinta power station of Taipower. Atmospheric Environment, 2005, vol. 39, no. 2, pp. 223–230.
Chien T. W., Hsueh H. T., Chu H. A feasibility study of a predictive emissions monitoring system applied to Taipower’s Nanpu and Hsinta power plants. Journal of Air Waste Manag. Assoc., 2010, vol. 60, no. 8, pp. 907–913. DOI: 10.3155/1047-3289.60.8.907.
Cooper D. A., Andreasson K. Predictive NOx emission monitoring on board a passenger ferry. Atmospheric Envi-ronment, 1999, vol. 33, no. 28, pp. 4637–4650.
Hung W. S. Y., Langenbacher F. PEMS: monitoring NOx emissions from gas turbines. ASME International Gas Turbine and Aeroengine Congress and Exposition, 1995, pp. 1–9. DOI: 10.1115/95-GT-415.
Kamas J., Keeler J. Predictive emissions monitoring systems: a low-cost alternative for emissions monitoring [in cement industry]. 1995 IEEE Cement Industry Technical Conference. 37th Conference Record, USA, 1995, pp. 497–509. DOI: 10.1109/CITCON.1995.514350.
Faravelli T., Bua L., Frassoldati A. A new procedure for predicting NOx emissions from furnaces. Computers & Chemical Engineering, 2001, vol. 25, iss. 4–6, pp. 613–618.
Harnevie H., Sarkoezi L., Trenkle S. Predictive emission monitoring system (PEMS) for emission control in bio-mass fired plants. Stiftelsen för värmeteknisk Forskning, 1996, no. SVF–575. 80 p.
Lee Y. H., Kim M., Han C. Application of multivariate statistical models to prediction of NOx emissions from complex industrial heater systems. Journal of Environmental Engineering, 2005, vol. 131, iss. 6, pp. 961–970.
Saiepour M., Schofield N., Leden B. Development and assessment of predictive emission monitoring systems (PEMS) models in the steel industry. Iron and Steel Technology Conference Proceedings. Ohio, 2006, vol. 2, 12 p.
Cuccu G., Danafar S., Cudre-Mauroux P. A data-driven approach to predict NOx-emissions of gas turbines. 2017 IEEE International Conference on Big Data, Boston, 2017, pp. 1283–1288. DOI: 10.1109/BigData.2017.8258056.
Vanderhaegen E., Deneve M., Laget H. Predictive emissions monitoring using a continuously updating neural network. ASME Turbo Expo 2010: Power for Land, Sea, and Air, 2010, vol. 2, pp. 769–775. DOI: 10.1115/GT2010-22899.
Skobelev D. O., Popov A. Yu., Ganyavin V. A. Approaches to building predictive emission control systems for modern industrial technological processes. Modeli, sistemy, seti v ekonomike, tekhnike, prirode i obshchestve, 2025, no. 3, pp. 46–64. (In Russ.). DOI: 10.21685/2227-8486-2025-3-4.
Seol Y., Lee S., Lee J. An interpretable time series forecasting model for predicting NOx emission concentration in ferroalloy electric arc furnace plants. Mathematics, 2024, vol. 12, no. 6, pp. 878. DOI: 10.3390/math12060878.
Malikov G. K., Lisenko V. G., Malikov K. Y., Viskanta R. A Mathematical modeling and validation study of NOx emissions in metal processing systems. ISIJ International, 2002, vol. 42, no. 10, pp. 1175–1181.
Chan E., Riley M., Thomson M. J., Evenson E. J. Nitrogen oxides (NOx) formation and control in an electric arc furnace (EAF): Analysis with measurements and computational fluid dynamics (CFD) modeling. ISIJ International, 2004, vol. 44, no. 2, pp. 429–438.
Skreiberg O., Hustad J. E., Garnaes E. I. Comparing empirical, statistical and neural network models calculating NOx emissions from gas turbines. International Journal of Energy for a Clean Environment, 2007, vol. 8, no. 3, pp. 221–237. DOI: 10.1615/INTERJENERCLEANENV.V8.I3.30.
Wang Y., Yang G., Xie R. An ensemble deep belief network model based on random subspace for NOx concen-tration prediction. ACS Omega, 2021, vol. 6, no. 11, pp. 7655–7668.
Kirschen M., Voj L., Pfeifer H. NOx emission from electric arc furnace – measurement and modelling. The Iron & Steel Technology Conference and Exposition, Association for Iron and Steel Technology, 2005, vol. 1, pp. 585–595.
Englund K. Quantification of the discharge of nitrous fumes from the steelplant of AB sandvik steel. Depart-ment of Chemical and Metallurgical Engineering, Lulea. 2002. Examensarbete, Luleåtekniska universitet, Civilingen-jörs programmet, 2002, vol. 246. pp. 1402–1617.
Bekker J. G., Craig I. K., Pistorius P. C. Model predictive control of an electric arc furnace off gas process. Con-trol Engineering Practice, 2000, vol. 8, no. 4, pp. 445–455. DOI: 10.1016/S0967-0661(99)00163-X.
Zhou H., Zhao J. P., Zheng L. G. Modeling NOx emissions from coal-fired utility boilers using support vector re-gression with ant colony optimization. Engineering Applications of Artificial Intelligence, 2012, vol. 25, no. 1, pp. 147–158. DOI: 10.1016/j.engappai.2011.08.005.
Iliyas S.A., Elshafei M., Habib M. A., Adeniran A. A. RBF neural network inferential sensor for process emission monitoring. Control Engineering Practice, 2013, vol. 21, no. 7, pp. 962–970. DOI: 10.1016/j.conengprac.2013.01.007.
Azzam M., Awad M., Zeaiter J. Application of evolutionary neural networks and support vector machines to model NOx emissions from gas turbines. Journal of Environmental Chemical Engineering, 2018, vol. 6, no. 1, pp. 1044–1052. DOI: 10.1016/j.jece.2018.01.020.
Korpela T., Kumpulainen P., Majanne Y. Indirect NOx emission monitoring in natural gas fired boilers. Control Engineering Practice, 2017, vol. 65, pp. 11–25. DOI: 10.1016/j.conengprac.2017.04.013.
Si M., Du K. Development of a predictive emissions model using a gradient boosting machine learning method. Environmental Technology & Innovation, 2020, vol. 20, 101028. DOI: 10.1016/j.eti.2020.101028.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Ferrous Metallurgy. Bulletin of Scientific , Technical and Economic Information
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
