Dizel Motorlarda Silindir İçi Basınç Tahmini için Veri Odaklı Makine Öğrenimi Modellerinin Performans Analizi

Year 2025, Volume: 29 Issue: 2, 335 - 349, 25.08.2025

Ahmet Karaoğlu , Hüseyin Söyler

https://doi.org/10.19113/sdufenbed.1623770

Abstract

Bu çalışma, dizel motorlarda silindir içi basınç tahmini için veri odaklı makine öğrenimi yaklaşımlarının performansını karşılaştırmayı amaçlamaktadır. Krank açısı ve yük değişkenlerine dayalı bir veri seti kullanılarak, Random Forest, Karar Ağacı ve XGBoost algoritmaları değerlendirilmiştir. Modellerin doğruluk oranları, işlem süreleri ve hata metrikleri detaylı bir şekilde analiz edilmiştir. Sonuçlar, Random Forest modelinin genelleme başarısı ve düşük hata oranları ile en iyi performansı sergilediğini göstermiştir. XGBoost modeli hızlı tahmin yeteneği ile dikkat çekerken, hafif doğruluk kayıpları sergilemiştir. Karar Ağacı modeli yüksek doğruluk sunmasına rağmen, genelleme yeteneği sınırlı kalmıştır. Bu modeller, emisyon kontrolü ve motor verimliliğini artırmaya yönelik çalışmalarda kullanılabilir güçlü tahmin araçları sunmaktadır. Elde edilen bulgular, veri odaklı yaklaşımların, dizel motorların performans ve emisyon analizinde güvenilir ve etkili bir alternatif sunduğunu ortaya koymaktadır.

Keywords

Dizel motor silindir içi basınç tahmini , XGBoost , Karar Ağacı , Random Forest , Makine öğrenmesi

Performance Analysis of Data-Driven Machine Learning Models for In-Cylinder Pressure Prediction in Diesel Engines

Abstract

This study aims to compare the performance of data-driven machine learning approaches for in-cylinder pressure prediction in diesel engines. Using a dataset based on crank angle and load variables, Random Forest, Decision Tree and XGBoost algorithms are evaluated. Accuracy rates, processing times and error metrics of the models are analyzed in detail. The results show that the Random Forest model performs the best with good generalization and low error rates. The XGBoost model was notable for its fast prediction capability, while exhibiting slight accuracy losses. Although the Decision Tree model offered high accuracy, its generalization ability was limited. These models provide powerful predictive tools that can be used in emission control and engine efficiency improvement studies. The findings suggest that data-driven approaches offer a reliable and effective alternative for performance and emission analysis of diesel engines.

Keywords

Diesel engine in-cylinder pressure prediction , Artificial neural networks , XGBoost , Random Forest , Machine learning

References

• [1] E. Alpteki̇n, “Metil ve Etil Ester Kullanılan Bir Common-Rail Dizel Motorda Performans, Yanma ve Enjeksiyon Karakteristiklerinin Karşılaştırılması”, Süleyman Demirel Üniversitesi Fen Bilim. Enstitüsü Derg., c. 21, sy 2, s. 578, Şub. 2017.
• [2] J. F. Dunne ve C. Bennett, “A crank-kinematics-based engine cylinder pressure reconstruction model”, Int. J. Engine Res., c. 21, sy 7, ss. 1147-1161, Eyl. 2020.
• [3] Y. Lee, S. Lee, ve K. Min, “Semi-empirical estimation model of in-cylinder pressure for compression ignition engines”, Proc. Inst. Mech. Eng. Part J. Automob. Eng., c. 234, sy 12, ss. 2862-2877, Eki. 2020.
• [4] W. Jeon vd., “Accelerometer-Based Robust Estimation of In-Cylinder Pressure for Cycle-to-Cycle Combustion Control”, IEEE Trans. Instrum. Meas., c. 72, ss. 1-13, 2023.
• [5] S. Kulah, A. Forrai, F. Rentmeester, T. Donkers, ve F. Willems, “Robust cylinder pressure estimation in heavy-duty diesel engines”, Int. J. Engine Res., c. 19, sy 2, ss. 179-188, Şub. 2018.
• [6] R. Ristow Hadlich, J. Loprete, ve D. Assanis, “A Deep Learning Approach to Predict In-Cylinder Pressure of a Compression Ignition Engine”, J. Eng. Gas Turbines Power, ss. 1-12, Oca. 2024.
• [7] C. Patil ve G. Theotokatos, “Comparative Analysis of Data-Driven Models for Marine Engine In-Cylind er Pressure Prediction”, Machines, c. 11, sy 10, s. 926, Eyl. 2023.
• [8] J. Castresana, G. Gabiña, L. Martin, ve Z. Uriondo, “Comparative performance and emissions assessments of a single-cylinder diesel engine using artificial neural network and thermodynamic simul ation”, Appl. Therm. Eng., c. 185, s. 116343, Şub. 2021.
• [9] B. Lee ve D. Jung, “Thermodynamics-based mean-value engine model with main and pilot injec tion sensitivity”, Proc. Inst. Mech. Eng. Part Journa Automob. Eng., c. 230, sy 13, ss. 1822-1834, Ağu. 2016.
• [10] H. Cho, B. Fulton, D. Upadhyay, T. Brewbaker, ve M. van Nieuwstadt, “In-cylinder pressure sensor–based NOx model for real-time a pplication in diesel engines”, Int. J. Engine Res., c. 19, sy 3, ss. 293-307, Nis. 2017.
• [11] J. Höller vd., “Parameter Estimation Strategies in Thermodynamics”, ChemEngineering, c. 3, sy 2, s. 56, Haz. 2019.
• [12] C. Mährle, S. Held, S. Huber, ve G. Wachtmeister, “A new method to determine the causes of deviation in cylinder pressure curves of motored reciprocating piston engines”, Int. J. Engine Res., c. 23, sy 2, ss. 243-261, Oca. 2021.
• [13] C. Jorques Moreno, O. Stenlaas, ve P. Tunestal, “Cylinder Pressure-Based Virtual Sensor for In-Cycle Pilot Mass Estimat ion”, SAE Int. J. Engines, c. 11, sy 6, ss. 1167-1182, Nis. 2018.
• [14] R. Ristow Hadlich, J. Loprete, ve D. Assanis, “A Deep Learning Approach to Predict In-Cylinder Pressure of a Compress ion Ignition Engine”, J. Eng. Gas Turbines Power, ss. 1-12, Oca. 2024.
• [15] C. Patil ve G. Theotokatos, “Comparative Analysis of Data-Driven Models for Marine Engine In-Cylind er Pressure Prediction”, Machines, c. 11, sy 10, s. 926, Eyl. 2023.
• [16] R. Ristow Hadlich, J. Loprete, ve D. Assanis, “A Deep Learning Approach to Predict In-Cylinder Pressure of a Compress ion Ignition Engine”, J. Eng. Gas Turbines Power, ss. 1-12, Oca. 2024.
• [17] E. Aslan ve Y. Özüpak, “Comparison of machine learning algorithms for automatic prediction of Alzheimer disease”, J. Chin. Med. Assoc., c. 88, sy 2, ss. 98-107, Şub. 2025.
• [18] E. Aslan, “Temperature Prediction and Performance Comparison of Permanent Magnet Synchronous Motors Using Different Machine Learning Techniques for Early Failure Detection”, Eksploat. Niezawodn. – Maint. Reliab., c. 27, sy 1, Ağu. 2024.
• [19] M. Sontheimer, A.-K. Singh, P. Verma, S.-Y. Chou, ve Y.-L. Kuo, “LSTM for Modeling of Cylinder Pressure in HCCI Engines at Different In take Temperatures via Time-Series Prediction”, Machines, c. 11, sy 10, s. 924, Eyl. 2023.
• [20] X. Jin, P. Cheng, W.-L. Chen, ve H. Li, “Prediction model of velocity field around circular cylinder over vario us Reynolds numbers by fusion convolutional neural networks based on p ressure on the cylinder”, Phys. Fluids, c. 30, sy 4, Nis. 2018.
• [21] S. A. Ali ve S. Saraswati, “Cycle-by-cycle estimation of cylinder pressure and indicated torque wa veform using crankshaft speed fluctuations”, Trans. Inst. Meas. Control, c. 37, sy 6, ss. 813-825, Eyl. 2014.
• [22] S. Trimby, J. F. Dunne, C. Bennett, ve D. Richardson, “Unified approach to engine cylinder pressure reconstruction using time -delay neural networks with crank kinematics or block vibration measur ements”, Int. J. Engine Res., c. 18, sy 3, ss. 256-272, Tem. 2016.
• [23] Q. Huang, J. Liu, C. Ulishney, ve C. E. Dumitrescu, “On the use of artificial neural networks to model the performance and emissions of a heavy-duty natural gas spark ignition engine”, Int. J. Engine Res., c. 23, sy 11, ss. 1879-1898, Tem. 2021.
• [24] F. Zhao ve D. L. S. Hung, “Applications of machine learning to the analysis of engine in-cylinder flow and thermal process: A review and outlook”, Appl. Therm. Eng., c. 220, s. 119633, Şub. 2023.
• [25] M. Henningsson, P. Tunestål, ve R. Johansson, “A Virtual Sensor for Predicting Diesel Engine Emissions from Cylinder Pressure Data”, IFAC Proc. Vol., c. 45, sy 30, ss. 424-431, 2012.
• [26] D. P. Viana vd., “Diesel Engine Fault Prediction Using Artificial Intelligence Regression Methods”, Machines, c. 11, sy 5, s. 530, May. 2023.
• [27] E. Tian, G. Lv, ve Z. Li, “Evaluation of emission of the hydrogen-enriched diesel engine through machine learning”, Energy, c. 307, s. 132303, Eki. 2024.
• [28] W. N. Wan Mansor, S. Abdullah, M. N. K. Jarkoni, J. S. Vaughn, ve D. B. Olsen, “Data on combustion, performance and emissions of a 6.8 L, 6-cylinder, Tier II diesel engine”, Data Brief, c. 33, s. 106580, Ara. 2020.
• [29] E. Aslan, “Prediction and Comparative Analysis of Emissions from Gas Turbines Using Random Search Optimization and Different MachineL earning Based Algorithms”, Bull. Pol. Acad. Sci. Tech. Sci., ss. 151956-151956, Eyl. 2024.
• [30] G. Reimer, “Online in-cylinder pressure and temperature prediction using a modeling approach: Masters Thesis”.
• [31] C. Patil ve G. Theotokatos, “Comparative Analysis of Data-Driven Models for Marine Engine In-Cylinder Pressure Prediction”, Machines, c. 11, sy 10, s. 926, Eyl. 2023.