Environmental Pollution Prediction of NOx by Predictive Modelling and Process Analysis in Natural Gas Turbine Power Plants

Document Type : Original Research Paper

Author

Applied Research and Innovation Services, Southern Alberta Institute of Technology, 1301 – 16 Avenue NW, Calgary, AB, Canada T2M 0L4

Abstract

The main objective of this paper is to propose K-Nearest-Neighbor (KNN) algorithm for predicting NOx emissions from natural gas electrical generation turbines. The process of producing electricity is dynamic and rapidly changing due to many factors such as weather and electrical grid requirements. Gas turbine equipment are also a dynamic part of the electricity generation since the equipment characteristics and thermodynamics behavior change as turbines age and equipment degrade gradually. Regular maintenance of turbines are also another dynamic part of the electrical generation process, affecting performance of equipment as parts and components may be upgraded over time. This analysis discovered using KNN, trained on a relatively small dataset produces the most accurate prediction rates in comparison with larger historical datasets. This observation can be explained as KNN finds the historical K nearest neighbor to the current input parameters and approximates a rated average of similar observations as prediction. This paper incorporates ambient weather conditions, electrical output as well as turbine performance factors to build a machine learning model predicting NOx emissions. The model can be used to optimize the operational processes for harmful emissions reduction and increasing overall operational efficiency. Latent algorithms such as Principle Component Algorithms (PCA) have been used for monitoring the equipment performance behavior change which deeply influences process paraments and consequently determines NOx emissions. Typical statistical methods of performance evaluations such as multivariate analysis, clustering and residual analysis have been used throughout the paper.
This paper incorporates ambient weather conditions, electrical output as well as turbine performance factors to build a machine learning model predicting NOx emissions. The model can be used to optimize the operational processes for harmful emissions reduction and increasing overall operational efficiency. Latent algorithms such as Principle Component Algorithms (PCA) have been used for monitoring the equipment performance behavior change which deeply influences process paraments and consequently determines NOx emissions. Typical statistical methods of performance evaluations such as multivariate analysis, clustering and residual analysis have been used throughout the paper.

Keywords


Bagheri, B., Ahmadi, H. and Labbafi, R. (2010). Application of data mining and feature extraction on intelligent fault diagnosis by Artificial Neural Network and k-nearest neighbor. The XIX International Conference on Electrical Machines - ICEM 2010, 1–7.
Biagioli, F. and Güthe, F. (2007). Effect of pressure and fuel–air unmixedness on NOx emissions from industrial gas turbine burners. Combustion and Flame, 151(1–2), 274–288.
Chen, X., Wang, P., Hao, Y. and Zhao, M. (2018). Evidential KNN-based condition monitoring and early warning method with applications in power plant. Neurocomputing, 315, 18–32.
Pollution 2021, 7(2): 481-494 493
Chien, T. W., Chu, H., Hsu, W. C., Tu, Y. Y., Tsai, H. S. and Chen, K. Y. (2005). A performance study of PEMS applied to the Hsinta power station of Taipower. Atmospheric Environment, 39(2), 223–230.
Chien, T. W., Hsueh, H. T., Chu, H., Hsu, W. C., Tu, Y. Y., Tsai, H. S. and Chen, K. Y. (2010). A Feasibility Study of a Predictive Emissions Monitoring System Applied to Taipower’s Nanpu and Hsinta Power Plants. Journal of the Air & Waste Management Association, 60(8), 907–913.
Cuccu, G., Danafar, S., Cudre-Mauroux, P., Gassner, M., Bernero, S. and Kryszczuk, K. (2017). A data-driven approach to predict NOx-emissions of gas turbines. 2017 IEEE International Conference on Big Data (Big Data), 1283–1288.
Environment and Climate Change Canada. (2017, November). Guidelines for the Reduction of Nitrogen Oxide Emissions from Natural Gas-fuelled Stationary Combustion Turbines.
European Environment Agency. (2013). Reducing air pollution from electricity-generating large combustion plants in the European Union: An assessment of potential emission reductions of NOX, SO2 and dust
European Environment Agency. (2019). Assessing the effectiveness of EU policy on large combustion plants in reducing air pollutant emissions
Fichet, V., Kanniche, M., Plion, P. and Gicquel, O. (2010). A reactor network model for predicting NOx emissions in gas turbines. Fuel, 89(9), 2202–2210.
Ge, Z., Song, Z., Ding, S. X. and Huang, B. (2017). Data Mining and Analytics in the Process Industry: The Role of Machine Learning. IEEE Access, 5, 20590–20616.
Hundi, P. and Shahsavari, R. (2020). Comparative studies among machine learning models for performance estimation and health monitoring of thermal power plants. Applied Energy, 265, 114775.
Kano, M., Hasebe, S., Hashimoto, I. and Ohno, H. (2004). Evolution of multivariate statistical process control: application of independent component analysis and external analysis. Computers & Chemical Engineering, 28(6–7), 1157–1166.
Kaya, H., Tufekci, P. and Gürgen, F. S. (2012, March). Local and Global Learning Methods for Predicting Power of a Combined Gas & Steam Turbine. International Conference on Emerging Trends in Computer and Electronics Engineering, 13–18.
Kaya, H., Tufekci, P. & Uzun, E. (2019). Predicting CO and NOx emissions from gas turbines: novel data and a benchmark PEMS. TURKISH JOURNAL OF ELECTRICAL ENGINEERING & COMPUTER SCIENCES, 27(6), 4783–4796.
Kourti, T. (2005). Application of latent variable methods to process control and multivariate statistical process control in industry. International Journal of Adaptive Control and Signal Processing, 19(4), 213–246.
Kuhn, M. and Johnson, K. (2013). Applied Predictive Modeling (1st ed. 2013, Corr. 2nd printing 2018 ed.). Springer.
Lee, Y.-H., Kim, M. and Han, C. (2005). Application of Multivariate Statistical Models to Prediction of NOx Emissions from Complex Industrial Heater Systems. Journal of Environmental Engineering, 131(6), 961–970.
Li, W., Zhang, C., Tsung, F. and Mei, Y. (2020). Nonparametric monitoring of multivariate data via KNN learning. International Journal of Production Research, 1–16.
Liukkonen, M. and Hiltunen, T. (2016). Monitoring and analysis of air emissions based on condition models derived from process history. Cogent Engineering, 3(1), 1174182.
Miletic, I., Quinn, S., Dudzic, M., Vaculik, V. and Champagne, M. (2004). An industrial perspective on implementing on-line applications of multivariate statistics. Journal of Process Control, 14(8), 821–836.
Pestov, V. (2013). Is the k-NN classifier in high dimensions affected by the curse of dimensionality? Computers & Mathematics with Applications, 65(10), 1427–1437.
Potter, M. and Somerton, C. (2019). Schaums Outline of Thermodynamics for Engineers, Fourth Edition (Schaum’s Outlines) (4th ed.). McGraw-Hill Education.
Poullikkas, A. (2005). An overview of current and future sustainable gas turbine technologies. Renewable and Sustainable Energy Reviews, 9(5), 409–443.
494 Rezazadeh
Proust, M. (2019). JMP® 15 Predictive and Specialized Modeling. SAS Institute Inc.
Qin, Z., Cen, C. and Guo, X. (2019). Prediction of Air Quality Based on KNN-LSTM. Journal of Physics: Conference Series, 1237, 042030.
SAS Institute Inc. (2014). SAS/STAT® 13.2 User’s Guide The VARCLUS Procedure.
Shakil, M., Elshafei, M., Habib, M. A. and Maleki, F. A. (2009). Soft sensor for and using dynamic neural networks. Computers & Electrical Engineering, 35(4), 578–586.
Si, M., Tarnoczi, T. J., Wiens, B. M. and Du, K. (2019). Development of Predictive Emissions Monitoring System Using Open Source Machine Learning Library – Keras: A Case Study on a Cogeneration Unit. IEEE Access, 7, 113463–113475.
Skiena, S. S. (2017). The Data Science Design Manual (Texts in Computer Science) (1st ed. 2017 ed.). Springer.
Smrekar, J., Potočnik, P. and Senegačnik, A. (2013). Multi-step-ahead prediction of NOx emissions for a coal-based boiler. Applied Energy, 106, 89–99.
Tüfekci, P. (2014). Prediction of full load electrical power output of a base load operated combined cycle power plant using machine learning methods. International Journal of Electrical Power & Energy Systems, 60, 126–140.
Yan, J., Meng, Y., Lu, L. and Li, L. (2017). Industrial Big Data in an Industry 4.0 Environment: Challenges, Schemes, and Applications for Predictive Maintenance. IEEE Access, 5, 23484–23491.