SavedAdd your journal

Journals / Hittite Journal of Science and Engineering / 2026 / Volume: 13 Issue: 3

Thermodynamic Consistency and Operational Logic Detection in Combined Cycle Power Plants through Explainable Gradient Boosting

Tuğba Tetik(İSTANBUL MEDENİYET ÜNİVERSİTESİ)0000-0002-0044-706X
Reha Avsar(İSTANBUL MEDENİYET ÜNİVERSİTESİ)0000-0001-9450-5728
Pages
—
Publication Date
2026-09-30
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

By mapping these data-driven inflection points to specific operational logic, this research demonstrates that a sensor-lean framework can achieve high maintenance resilience and function as a reliable data-driven operational surrogate for modern energy engineering.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Accurate forecasting of net electrical power output (PE) in Combined Cycle Power Plants (CCPP) is a requisite for grid stability and strategic bidding in deregulated energy markets.
📊
Key Findings & Data:Utilizing a benchmark dataset of 9,568 operational hours, with an 80/20 train/test split and 5-fold cross-validation scheme, the optimized CatBoost architecture achieved the highest predictive fidelity, yielding a mean Coefficient of Determination (R2) of 0.969 and an RMSE of 2.97 MW during cross-validation.
💡
Conclusion & Impact:By mapping these data-driven inflection points to specific operational logic, this research demonstrates that a sensor-lean framework can achieve high maintenance resilience and function as a reliable data-driven operational surrogate for modern energy engineering.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Combined Cycle Power Plant · Real-Time Control · Gradient Boosting

Abstract

Accurate forecasting of net electrical power output (PE) in Combined Cycle Power Plants (CCPP) is a requisite for grid stability and strategic bidding in deregulated energy markets. While Gradient Boosting Decision Tree (GBDT) architectures offer superior predictive precision compared to traditional physical simulations, their black-box opacity often hinders industrial adoption. This study presents a physics-consistency forecasting framework that combines a comparative stress-test of XGBoost, LightGBM, and CatBoost, with post-hoc SHapley Additive exPlanations (SHAP) analysis for thermodynamic validation. Utilizing a benchmark dataset of 9,568 operational hours, with an 80/20 train/test split and 5-fold cross-validation scheme, the optimized CatBoost architecture achieved the highest predictive fidelity, yielding a mean Coefficient of Determination (R2) of 0.969 and an RMSE of 2.97 MW during cross-validation. Furthermore, rigorous computational latency benchmarking (10,000 iterations) and jitter analysis confirmed CatBoost as the optimal engine for real-time Automatic Generation Control (AGC), demonstrating a sub-millisecond inference latency of 999 µs per sample. A pivotal contribution of this research is the unsupervised identification of a 24.6°C thermodynamic inflection point via bootstrap SHAP interaction analysis. This data-driven threshold is consistent with the onset of psychrometric performance degradation and the expected activation logic of auxiliary inlet air cooling systems, suggesting the model has internalized Brayton cycle thermodynamics without explicit physics programming. By mapping these data-driven inflection points to specific operational logic, this research demonstrates that a sensor-lean framework can achieve high maintenance resilience and function as a reliable data-driven operational surrogate for modern energy engineering.

🔖

Cite This Article

Tetik, T., Avsar, R. (2026). Thermodynamic Consistency and Operational Logic Detection in Combined Cycle Power Plants through Explainable Gradient Boosting. Hittite Journal of Science and Engineering, Volume: 13 Issue: 3. https://doi.org/10.17350/HJSE19030000382

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (34)

  1. Bhatia K, Mittal R, Varanasi J, Tripathi MM. An ensemble approach for electricity price forecasting in markets with renewable energy resources. Util Policy. 2021;70:101185.
  2. Haque SJ, Uddin KI, Nakib AM, Waleed K, Hossain MZ, Rakib NA. Enhanced Prediction of Power Plant Energy Output Using Advanced Ensemble Machine Learning Models. Eur J Appl Sci Eng Technol. 2025;3(5):105–15.
  3. Liu Z, Karimi IA. Gas turbine performance prediction via machine learning. Energy. 2020;192:116627.
  4. Tüfekci P. Prediction of full load electrical power output of a base load operated combined cycle power plant using machine learning methods. Int J Electr Power Energy Syst. 2014;60:126–40.
  5. Kaya H, Tüfekci P, Gürgen SF. Local and Global Learning Methods for Predicting Power of a Combined Gas & Steam Turbine. In: International Conference on Emerging Trends in Computer and Electronics Engineering (ICETCEE 2012). Dubai; 2012. p. 13–8.
  6. Siddiqui R, Anwar H, Ullah F, Ullah R, Rehman MA, Jan N, et al. Power Prediction of Combined Cycle Power Plant (CCPP) Using Machine Learning Algorithm-Based Paradigm. Wirel Commun Mob Comput. 2021;
  7. Dursun M, Ersöz B, Oyucu S, Aköz A. A Comparative Analysis of Boosting-Based Ensemble Learning Models and Stacking Ensemble Approach for Electricity Generation Forecasting in Thermal Power Plants. In: 12th International Conference on Electrical and Electronics Engineering. 2025. p. 118–22.
  8. Tetik T, Cihan P. Machine learning approaches to ground motion models for duration and velocity-based intensity measures in Türkiye with magnitude-dependent performance assessment. Eng Appl Artif Intell. 2026;168:114053.
  9. Tetik T, Cihan P. Non-parametric ground motion duration and energy-based intensity measure models: A novel stacking ensemble model using regional seismic data. Soil Dyn Earthq Eng. 2026;203:110135.
  10. Guo B, Yang B, Shi W, Yang F, Wang D, Wang S. CCPP Power Prediction Using CatBoost with Domain Knowledge and Recursive Feature Elimination. Energies. 2025;18(16):4272.
  11. Qu Z, Xu J, Wang Z, Chi R, Liu H. Prediction of electricity generation from a combined cycle power plant based on a stacking ensemble and its hyperparameter optimization with a grid-search method. Energy. 2021;227:120309.
  12. Hong CW, Kim J. Exhaust Temperature Prediction for Gas Turbine Performance Estimation by Using Deep Learning. J Electr Eng Technol. 2023;18(4):3117–25.
  13. Khwaja AS, Anpalagan A, Naeem M, Venkatesh B. Joint bagged-boosted artificial neural networks: Using ensemble machine learning to improve short-term electricity load forecasting. Electr Power Syst Res. 2020;179:106080.
  14. Mast J, Radle S, Gerlach J, Bringmann O. Exploration of machine learning methods for predicting the operation schedule of a combined heat and power plant. 2019 IEEE Milan PowerTech, PowerTech 2019. 2019;
  15. Aslan A, Büyükköse AO. The Effects of Increasing Ambient Temperature and Sea Surface Temperature Due to Global Warming on Combined Cycle Power Plant. Sustainability. 2025;17(10):4605.
  16. Aslan A, Büyükköse AO. Comparative Performance Analysis of Machine Learning-Based Annual and Seasonal Approaches for Power Output Prediction in Combined Cycle Power Plants. Energies. 2025;18(19):1–28.
  17. Machlev R, Heistrene L, Perl M, Levy KY, Belikov J, Mannor S, et al. Explainable Artificial Intelligence (XAI) techniques for energy and power systems: Review, challenges and opportunities. Energy AI. 2022;9:100169.
  18. Pelekis S, Karakolis E, Lampropoulos G, Mouzakitis S, Markaki O, Ntanos C, et al. Trustworthy Artificial Intelligence in the Energy Sector: Landscape Analysis and Evaluation Framework. Proc 30th ICE IEEE/ITMC Conf Eng Technol Innov Digit Transform Eng Technol Innov ICE 2024. 2024;
  19. Devanathan B, Varshitha KJ, Kumar LP, Lakshmanan SA, Prakash NK. Explainable AI Framework Using XGBoost With SHAP and LIME for Multi-Scale Household Energy Forecasting. IEEE Access. 2025;13:149750–64.
  20. Matrenin P V., Stepanova AI. Enhancing the interpretability of electricity consumption forecasting models for mining enterprises using SHapley Additive exPlanations. J Min Inst. 2025;271:154–67.
  21. Arrieta AB, Díaz-Rodríguez N, Del Ser J, Bennetot A, Tabik S, Barbado A, et al. Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Inf Fusion. 2020;58(2020):82–115.
  22. Tetik T. Feature engineering and explainable artificial intelligence for state of health estimation of Lithium-ion batteries. J Energy Storage. 2026;144:119873.
  23. van Zyl C, Ye X, Naidoo R. Harnessing eXplainable artificial intelligence for feature selection in time series energy forecasting: A comparative analysis of Grad-CAM and SHAP. Appl Energy. 2024;353:122079.
  24. Singla A. Explainable Artificial Intelligence for Site Energy Usage Intensity Prediction Master’s Thesis. 2024.
  25. Daif N, Di Nunno F, Granata F, Difi S, Kisi O, Heddam S, et al. Forecasting maximal and minimal air temperatures using explainable machine learning: Shapley additive explanation versus local interpretable model-agnostic explanations. Stoch Environ Res Risk Assess. 2025;39(6):2551–81.
  26. Joy RA. An Interpretable Catboost Model to Predict the Power of Combined Cycle Power Plants. In: 2021 International Conference on Information Technology, ICIT 2021 - Proceedings. IEEE; 2021. p. 435–9.
  27. Zhang J, Zhang M, Yang J, Zheng X. Prediction of electricity load generated by Combined Cycle Power Plants using integration of machine learning methods and HGS algorithm. Comput Electr Eng. 2024;120:109644.
  28. Chen T, Guestrin C. XGBoost: A scalable tree boosting system. In: Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. California, San Francisco, USA; 2016. p. 785–94.
  29. Ke G, Meng Q, Finley T, Wang T, Chen W, Ma W, et al. LightGBM: A Highly Efficient Gradient Boosting Decision Tree. In: 31st Conference on Neural Information Processing Systems (NIPS 2017). Long Beach, CA; 2017. p. 3149–57.
  30. Prokhorenkova L, Gusev G, Vorobev A, Dorogush AV, Gulin A. Catboost: Unbiased boosting with categorical features. In: 32nd Conference on Neural Information Processing Systems. Montréal, Canada; 2018.
  31. Lundberg SM, Lee SI. A unified approach to interpreting model predictions. In: 31st Conference on Neural Information Processing Systems (NIPS2017). Long Beach, CA, USA; 2017.
  32. Bevrani H. Robust Power System Frequency Control. 2nd ed.; 2009.
  33. Grieves M, Vickers J. Digital Twin: Mitigating Unpredictable, Undesirable Emergent Behavior in Complex Systems. Transdisciplinary Perspectives on Complex Systems: New Findings and Approaches. Springer International Publishing; 2017. 85–113 p.
  34. Tufekci P, Kaya H. Combined Cycle Power Plant [dataset]. 2014. UCI Machine Learning Repository. Available from: https://doi.org/10.24432/C5002N.