International Journal of Emerging Research in Science, Engineering, and Management
Vol. 1, Issue 3, pp. 01-05, September 2025.
This work is licensed under a Creative Commons Attribution 4.0 International License.
Optimization of Hybrid XGBoost–GPR Models for Predictive Maintenance in Gas Turbine Systems
D. Janani
E. Sasikala Reddy
Assistant Professor, Department of CSE, Siddharth Institute of Science and Technology, Puttur, India.
Associate Professor, Department of ECE, Gokula Krishna College of Engineering, Sullurpet, India.
Abstract: Gas turbines are widely deployed in power generation and industrial applications, but their complex operating environments make them prone to performance degradation and unexpected failures. Predictive maintenance is crucial for ensuring reliable operation; however, existing models face challenges related to scalability, interpretability, and representation of uncertainty. This study proposes a hybrid predictive framework integrating Extreme Gradient Boosting (XGBoost) with Gaussian Process Regression (GPR) for gas turbine health monitoring. XGBoost captures nonlinear dependencies among operational parameters, while GPR models the residuals to provide quantification of uncertainty. Simulation experiments using a 25,000-point SCADA dataset demonstrated that the hybrid framework achieved superior predictive performance, reducing RMSE by 18.7% compared to standalone XGBoost and by 31.7% compared to Random Forest regression, with an overall R² of 0.97. Furthermore, the model delivered reliable uncertainty estimates with a 94.3% coverage probability at the 95% confidence level. These results highlight the proposed hybrid XGBoost–GPR model as a promising approach for predictive maintenance, enabling proactive fault detection, reducing false alarms, and supporting cost-effective, risk-aware decision-making in gas turbine operations.
Keywords: Predictive maintenance, Gas turbines, Hybrid modelling, Extreme Gradient Boosting (XGBoost), Gaussian Process Regression (GPR), Uncertainty quantification, SCADA data analysis.
References:
- Y. Zheng, X. Zhou, J. Yu, X. Xue, X. Wang, and X. Tu, “Predictive Analytics for Sustainable Energy: An in-depth assessment of novel stacking regressor model in the Off-Grid Hybrid Renewable Energy systems,” Energy, p. 135916, Apr. 2025, doi: 10.1016/j.energy.2025.135916.
- X. Liu, “Proposing an innovative model for solar irradiance and wind speed forecasting,” Applied Thermal Engineering, p. 125224, Dec. 2024, doi: 10.1016/j.applthermaleng.2024.125224.
- L. Sun, T. Liu, Y. Xie, D. Zhang, and X. Xia, “Real-time power prediction approach for turbine using deep learning techniques,” Energy, vol. 233, p. 121130, Jun. 2021, doi: 10.1016/j.energy.2021.121130.
- S. Zampini, G. Parodi, L. Oneto, A. Coraddu, and D. Anguita, “A review on full-, zero-, and partial-knowledge based predictive models for industrial applications,” Information Fusion, p. 102996, Feb. 2025, doi: 10.1016/j.inffus.2025.102996.
- M. D. Mukelabai, E. R. Barbour, and R. E. Blanchard, “Modeling and optimization of renewable hydrogen systems: A systematic methodological review and machine learning integration,” Energy and AI, p. 100455, Nov. 2024, doi: 10.1016/j.egyai.2024.100455.
- Z. Allal, H. N. Noura, O. Salman, and K. Chahine, “Machine learning solutions for renewable energy systems: Applications, challenges, limitations, and future directions,” Journal of Environmental Management, vol. 354, p. 120392, Feb. 2024, doi: 10.1016/j.jenvman.2024.120392.
- D. Zemin, L. Yueming, Y. Zhicheng, Y. Youhong, and L. Yongbao, “Coordinated control strategy of engine-grid-load-storage for shipboard micro gas turbine DC power generation system: A review,” Journal of Energy Storage, vol. 134, p. 118205, Aug. 2025, doi: 10.1016/j.est.2025.118205.
- A. Harutyunyan, K. Badyda, and Ł. Szablowski, “Energy and exergy analysis of complex gas turbines systems powered by a mixture of hydrogen and methane,” International Journal of Hydrogen Energy, Feb. 2025, doi: 10.1016/j.ijhydene.2025.02.378.
- R. S. Aweid, A. N. Mustafa, O. M. Ali, and B. M. Ali, “Thermal, environmental, and economic analysis of the gas turbine fogging system,” Results in Engineering, p. 103952, Jan. 2025, doi: 10.1016/j.rineng.2025.103952.
- Y. Li et al., “A large-scale power-to-H2-to-power system adopting hydrogen mixed gas turbine for wind accommodation: Process modeling, optimal dispatch and economic feasibility analysis,” International Journal of Hydrogen Energy, vol. 145, pp. 345–357, Jun. 2025, doi: 10.1016/j.ijhydene.2025.06.034.
- Y. Zhou, W. Chen, D. Wang, and R. Zhu, “Dynamics modeling and impact resistance study of the marine gas turbine isolation system,” Ocean Engineering, vol. 331, p. 121225, Apr. 2025, doi: 10.1016/j.oceaneng.2025.121225.
- R. González-Almenara, L. García-Rodríguez, and D. Sánchez, “Experimental assessment of a zero liquid discharge system driven by a micro gas turbine,” Journal of Water Process Engineering, vol. 76, p. 108058, Jun. 2025, doi: 10.1016/j.jwpe.2025.108058.