Estimation of Current and Voltage Values Generated from a Thermoelectric Generator Mounted on Automobile Exhaust System by Machine Learning Algorithms: A Comparative Study

Abstract

Electricity is one of the most important sources of energy. Many devices need electrical energy to operate. In addition to the production of electrical energy from renewable sources, the fact that it can be produced from waste heat sources will increase efficiency. As in many systems, it is possible to generate electricity by using thermoelectric generators (TEGs) on the waste heat systems of vehicles using internal combustion engines. Thanks to the electricity obtained from waste heat systems, the load on the alternators and batteries in the vehicles is reduced, thus increasing their service life. In addition, since the charging time of the vehicle battery is reduced, fuel savings can be achieved. Therefore, making electricity generation predictions using machine learning algorithms in internal combustion engines will make a great contribution to the initial project planning phase of the design of automobile systems. Nowadays, research on waste heat energy recovery from automobile exhaust with TEGs using machine learning is a new topic. In this study, a data set containing the attributes of 2692 current and voltage values obtained from a thermoelectric generator on an automobile exhaust system was used. Adaboost and Random Forest machine learning algorithms were used in the estimation process of the designed model. The most successful result was achieved when estimating the current with the Adaboost algorithm. In this study, it has been shown that with the proposed model, electrical energy production estimation can be made over the waste heat sources of different systems.

Keywords

• Thermoelectric generator
• waste heat source
• machine learning
• estimation
• performance measurement

Otomobil Egzoz Sistemine Monte Edilen Termoelektrik Jeneratörden Üretilen Akım ve Gerilim Değerlerinin Makine Öğrenmesi Algoritmaları ile Tahmini: Karşılaştırmalı Bir Çalışma

Abstract

Elektrik, en önemli enerji kaynaklarından biridir. Birçok cihaz çalışmak için elektrik enerjisine ihtiyaç duyar. Yenilenebilir kaynaklardan elektrik enerjisi üretmenin yanı sıra atık ısı kaynaklarından da elektrik enerjisi üretilebilmesi verimliliği artıracaktır. Birçok sistemde olduğu gibi, içten yanmalı motor kullanan araçların atık ısı sistemlerinde termoelektrik jeneratörler (TEG) kullanarak elektrik üretmek mümkündür. Atık ısı sistemlerinden elde edilen elektrik sayesinde araçlardaki alternatör ve akülerin yükü azaltılarak ömürleri uzatılır. Ayrıca araç aküsünün şarj süresi kısaldığı için yakıt tasarrufu sağlanır. Bu nedenle içten yanmalı motorlarda makine öğrenmesi algoritmaları kullanarak elektrik üretimi tahminleri yapmak, otomobil sistemlerinin tasarımında ilk proje planlama aşamasına büyük katkı sağlayacaktır. Günümüzde, makine öğrenmesi kullanılarak otomobil egzozundan atık ısı enerjisinin TEG’ler ile geri kazanımı üzerine yapılan araştırmalar yeni bir konudur. Bu çalışmada, bir otomobil egzoz sistemi üzerinde bulunan bir termoelektrik jeneratörden elde edilen 2692 akım ve gerilim değerlerinin özniteliklerini içeren bir veri seti kullanılmıştır. Tasarlanan modelin tahmin sürecinde Adaboost ve Rastgele Orman makine öğrenme algoritmaları kullanılmıştır. Akımın tahmininde Adaboost algoritması kullanıldığında en başarılı sonuç elde edilmiştir. Bu çalışmada, önerilen model ile farklı sistemlerin atık ısı kaynakları üzerinden elektrik enerjisi üretim tahmini yapılabildiği gösterilmiştir.

Keywords

• Termoelektrik jeneratör
• atık ısı kaynağı
• makine öğrenmesi
• tahmin
• performans ölçümü

References

1. [1] Burnete, N. V., Mariasiu, F., Depcik, C., Chiriac, A., Cormos, C. C., Vekas, D. I. and Lucaciu, C. R. Review of thermoelectric generation for internal combustion engine waste heat recovery. Progress in Energy and Combustion Science, 2022, 91, pp. 101009.
2. [2] Kunt, M. A. A design of a liquid cooling thermoelectric generator system for the exhaust systems of internal combustion engines and experimental study on the effect of refrigerant fluid quantity on recovery performance. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, 2019, 25, pp. 7–12.
3. [3] Kunt, M. A. An Experimental Investigation of Exhaust Waste Heat Recycling by Thermoelectric Generators Under Different Thermal Conditions for Internal Combustion Engines. Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering, 2017, 232, pp. 1–6.
4. [4] Kunt, M. A. A comparative survey on recovery performance of thermo-electric generator recovery system with air cooling during warming process of internal combustion gasoline engines. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, 2018, 24, pp. 605–609.
5. [5] Albatati, F. and Attar, A. Analytical and Experimental Study of Thermoelectric Generator (TEG) System for Automotive Exhaust Waste Heat Recovery. Energies, 2021, 14, pp. 204.
6. [6] Li, X., Xie, C., Quan, S., Zhao, J., Tang, Y., Chen, Y. and Xu, J. Optimization of Thermoelectric Modules’ Number and Distribution Pattern in an Automotive Exhaust Thermoelectric Generator. IEEE Access, 2019, 99, pp. 1–1.
7. [7] Duzgun, M. and Karabulut, H. Thermal Performance Analysis of a Stirling Engine Energized with Exhaust Gas of a Diesel Engine. Journal of Thermal Science and Technology, 2021, 41, pp. 249–263.
8. [8] Tulkinov, S. Forecast of Electricity Production from Coal and Renewable Sources in Major European Economies. Research Square, 2023, pp. 1–32.

9. [9] Angeline, A. A., Asirvatham, L. G., Hemanth, D. J., Kumar, R. S., and Rajesh, R. Performance prediction of hybrid thermoelectric generator with high accuracy using artificial neural networks. Sustainable Energy Technologies and Assessments, 2019, 33, pp. 53–60.
10. [10] Ledmaoui, Y., El Maghraoui, A. and El Aroussi, M. Forecasting solar energy production: A comparative study of machine learning algorithms. Energy Reports, 2023, 10, pp. 1004–1012.
11. [11] Tanveer, A. and Huanxin, C. Deep learning for multi-scale smart energy forecasting. Energy, 2019, 175, pp. 98–112.
12. [12] Sebetci, Ö., Şimşek, M and Yılmaz, İ. Classification of Cell Line Halm Machine Data in Solar Energy Panel Production Factories Using Artificial Intelligence Models. El-Cezeri Science and Engineering Journal, 2025, 12, pp. 44-53.
13. [13] Saleh, H. Wind Power Prediction Based on Wind Velocity Variable and Polynomial Regression Method., El-Cezeri Science and Engineering Journal, 2025, 12, pp. 274-282.
14. [14] Chang, G. W. and Lu, H. J. Integrating grey data preprocessor and deep belief network for day-ahead PV power output forecast. IEEE Transactions on Sustainable Energy, 2019, 99, pp. 185–194.
15. [15] Kadar, P. and Lovassy, R. Spatial load forecast for Electric Vehicles. In Proceedings of the 4th IEEE International Symposium on Logistics and Industrial Informatics (LINDI), Smolenice, Slovakia, September 5–7, 2012, pp. 163–168.
16. [16] MdShahiduzzaman, K., Jamal, M. N. and IbnNawab, R. Renewable Energy Production Forecasting: A Comparative Machine Learning Analysis. International Journal of Engineering and Advanced Technology, 2021, 10, pp. 11–18.
17. [17] Kishore, S., Patel, R., Singh, A., and Verma, K. Temperature prediction algorithms using machine learning for electric vehicles. In Proceedings of the IEEE 3rd International Conference on Sustainable Energy and Future Electric Transportation (SEFET), August 2023, pp. 1–6.
18. [18] Baran, I., Yilmaz, T., Kaya, M., and Demir, S. Using numerical weather forecast to predict power losses on transmission lines. In Proceedings of the 4th International Symposium on Electrical and Electronics Engineering (ISEEE), October 2013, pp. 1–8.
19. [19] Ullah, A., Khan, M., Ahmed, S., Ali, R., and Hussain, T. AlexNet, AdaBoost and artificial bee colony based hybrid model for electricity theft detection in smart grids. IEEE Access, 2022, 10, pp. 18681–18694.
20. [20] Chang, L., Wang, Y., Zhao, H., and Chen, Q. Application of an improved BP-AdaBoost model in semiconductor quality prediction. In Proceedings of the IEEE International Symposium on Predictive Control of Electrical Drives and Power Electronics (PRECEDE), May–June 2019, pp. 1–4.
21. [21] Khudhair, I., Al-Mashhadani, M., Jasim, H., and Kareem, A. Data mining and analysis for predicting electrical energy consumption. Bulletin of Electrical Engineering and Informatics, 2023, 12, pp. 997–1006.
22. [22] Adhya, D., Saha, P., Roy, A., and Choudhury, S. Machine learning application for prediction of EV charging demand for the scenario of Agartala, India. In Proceedings of the 4th International Conference on Energy, Power and Environment (ICEPE), April–May 2022, pp. 1–5.
23. [23] Rajagopalan, P., Mehta, S., Kumar, V., and Sharma, R. Predicting excess energy and estimating users for vehicle-to-grid (V2G) services using machine learning. In Proceedings of the IEEE 14th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference (UEMCON), October 2023, pp. 0507–0515.
24. [24] Nti, I., Adekoya, A., Nyarko-Boateng, O., Mensah, J., and Owusu, K. Electricity load forecasting: a systematic review. Journal of Electrical Systems and Information Technology, 2020, 7, pp. 13.
25. [25] Parhizkar, T. Random Forest method for energy consumption prediction. Journal of Cleaner Production, 2021, 279, pp. 1–17.
26. [26] Wang, X., Liu, Y., Zhang, H., and Chen, L. Forecast of electric vehicle ownership based on MIFS-AdaBoost model. In Proceedings of the IEEE 4th International Conference on Automation, Electronics and Electrical Engineering (AUTEEE), November 2021, pp. 4–8.
27. [27] Mary, J. Prediction and Comparison using AdaBoost and ML Algorithms with Autistic Children Dataset. International Journal of Engineering Research, 2020, 9, pp. 563–567.
28. [28] Kempf, N. and Zhang, Y. Design and optimization of automotive thermoelectric generators for maximum fuel efficiency improvement. Energy Conversion and Management, 2016, 121, pp. 224–231.
29. [29] Agudelo, A. F., García-Contreras, R., Agudelo, J. R., López, J. A., and Torres, M. Potential for exhaust gas energy recovery in a diesel passenger car under European driving cycle. Applied Energy, 2016, 174, pp. 201–212.
30. [30] Lia, B., Huanga, K., Yana, Y., Chen, L., and Wu, Z. Heat transfer enhancement of a modularised thermoelectric power generator for passenger vehicles. Applied Energy, 2017, 205, pp. 868–879.
31. [31] Massaguer, A. Transient behavior under a normalized driving cycle of an automotive thermoelectric generator. Applied Energy, 2017, 206, pp. 1282– 1296.
32. [32] Cao, Q., Li, H., Zhang, Y., and Wang, J. Performance enhancement of heat pipes assisted thermoelectric generator for automobile exhaust heat recovery. Applied Thermal Engineering, 2018, 130, pp. 1472–1479.
33. [33] Burnete, N. V., Moldovan, A., Mureșan, R., and Pop, C. Simulink model of a thermoelectric generator for vehicle waste heat recovery. Applied Sciences, 2021, 11, pp. 1340.
34. [34] Karabektaş, M., Yılmaz, H., Demir, A., and Kaya, S. Egzoz atık ısısından termoelektrik modülle enerji üreten sistem tasarımı ve optimizasyonu. In Proceedings of the 6th International Symposium on Innovative Technologies in Engineering and Science, November 2018, pp. 358–367.
35. [35] Temizer, İ., Yıldız, M., Aksoy, E., and Demirtaş, H. Dizel motor egzoz sistemi için termoelektrik jeneratör uygulaması ve akış analizi. Afyon Kocatepe Üniversitesi Fen ve Mühendislik Bilimleri Dergisi, 2016, 16, pp. 431–445.
36. [36] Al-Ghzawi, M. and El-Rayes, K. Machine learning for predicting the impact of construction activities on air traffic operations during airport expansion projects. Automation in Construction, 2024, 158, pp. 105189.
37. [37] Çelik, A. Predicting diagnosis of COVID-19 disease with AdaBoost and Naive Bayes machine learning algorithms. Journal of Engineering Sciences and Design, 2022, 10, pp. 1212–1221.
38. [38] Imtinan Uddin, J. M. Depression risk prediction among tech employees in Bangladesh using Adaboosted Decision Tree. In Proceedings of the IEEE International Women in Engineering (WIE) Conference on Electrical and Computer Engineering (WIECON-ECE), December 2020, pp. 135–138.
39. [39] Acılar, A. M. ADABOOST.R2 regresyon algoritması ile konutların ısıtma ve soğutma yüklerinin tahmin edilmesi. EJONS International Journal on Mathematics and Engineering Sciences, 2021, 4, pp. 1–10.
40. [40] Li, Y., Zoub, C., Berecibar, M., Nanini-Maury, E., Omar, N., and Van den Bossche, P. Random Forest regression for online capacity estimation of lithium-ion batteries. Applied Energy, 2018, 232, pp. 197–210.
41. [41] Lahouar, A. and Slama, J. B. H. Hour-ahead wind power forecast based on random forests. Renewable Energy, 2017, 41, pp. 529–541.
42. [42] Opara, J., Aimufua, G. I. O., Abdullahi, M. U., Ibrahim, S., and Musa, A. Churn prediction in telecommunication industry: A comparative analysis of boosting algorithms. Dutse Journal of Pure and Applied Sciences, 2024, 10, pp. 313–324.
43. [43] Sulaiman, M. H. and Mustaffa, Z. State of charge estimation for electric vehicles using random forest. Green Energy and Intel ligent Transportation, 2024, 3, pp. 100177.
44. [44] Kong, D., Liu, Y., Zhang, H., and Wang, X. Forecasting urban carbon emissions using an Adaboost-STIRPAT model. Frontiers in Environmental Science, 2023, 11, pp. 1–13.
45. [45] Freund, Y. and Schapire, R. A decision-theoretic generalisation of on-line learning and an application of boosting. Journal of Computer and System Sciences, 1997, 55, pp. 119–139.
46. [46] Solomatine, D. P. and Shrestha, D. L. AdaBoost.RT: A boosting algorithm for regression problems. In Proceedings of the 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541), 2, 2004, pp. 1163–1168.
47. [47] Ho, T. K. Random decision forests. In Proceedings of the 3rd International Conference on Document Analysis and Recognition, August 1995, pp. 278– 282.
48. [48] Breiman, L. Random forests. Machine Learning, 2001, 45, pp. 5–32.
49. [49] Han, S., Kim, H., and Lee, Y. S. Double random forest. Machine Learning, 2020, 109, pp. 1569–1586.
50. [50] Eskicioğlu Ö.C., Işık A.H. and Sevli O. Machine Learning Detection of Collision-Risk Asteroids. El-Cezerî Journal of Science and Engineering, 2022, 9, pp. 1431- 1449.
51. [51] Schneider, P., Wingerath, W., Gossen, F., and Ritter, N. Anomaly detection and complex event processing over IoT data streams. In Anomaly Detection and Complex Event Processing over IoT Data Streams. Elsevier Academic Press, 2022, pp. 49–66.
52. [52] Singh, M., Sharma, A., Gupta, R., and Patel, K. Artificial intelligence and machine learning for EDGE computing. In Artificial Intelligence and Machine Learning for EDGE Computing. Elsevier Academic Press, 2022, pp. 235–254.
53. [53] Cetin, M., Urkan, O. D., Hekim, M., and Yilmaz, A. Power generation prediction of a geothermal-thermoelectric hybrid system using intelligent models. Geothermics, 2024, 118, pp. 102911.
54. [54] Luo, D., Liu, Z., Yan, Y., and Chen, H. Recent advances in modeling and simulation of thermoelectric power generation. Energy Conversion and Management, 2022, 273, pp. 116389.
55. [55] Belovski, I., Petrov, D., Nikolov, S., and Georgiev, K. Thermoelectric generator power prediction based on artificial neural network. In Proceedings of the 20th International Symposium on Electrical Apparatus and Technologies (SIELA), June 2018, pp. 1–4.
56. [56] Ozbektas, S., Kaleli, A., and Sungur, B. Prediction of the effect of load resistance and heat input on the performance of thermoelectric generator using numerical and artificial neural network models. Applied Thermal Engineering, 2024, 249, pp. 123417.
57. [57] Celik, A., Kunt, M. A., and Gunes, H. Prediction of electric power performance of the exhaust waste heat recovery system of an automobile with thermoelectrical generator under real driving conditions by means of machine learning algorithms. Proceedings of the Institution of Mechanical Engineers, Part E: Journal of Process Mechanical Engineering, 2024, 238, pp. 1873–1883.