Research Article
BibTex RIS Cite

Elektrikli Araçların Enerji Tüketimini Tahmin Etmede Makine Öğrenimi Algoritmalarının Performanslarının Karşılaştırılması

Year 2024, , 518 - 530, 01.06.2024
https://doi.org/10.21597/jist.1448216

Abstract

Ulaştırma sektörünün hızlı büyümesi ve buna bağlı emisyonlar, çevresel sürdürülebilirliğin sağlanması önem teşkil etmektedir. Bu nedenle, ulaştırma emisyonlarının türetilme faktörünün anlaşılması son derece önemlidir. Artan ulaşım emisyonları karşısında elektrikli araçların (EA) rolü kullanımının yaygınlaştırılması önemlidir. Elektrikli araçlar düşük karbon ekonomisine ve sürdürülebilir çevreye giden yolu açmaktadır. Elektrikli araçların başarılı bir şekilde yaygınlaştırılması, büyük ölçüde enerji tüketimini verimli ve güvenilir bir şekilde tahmin edebilen enerji tüketim modellerine dayanmaktadır. Elektrikli araçların enerji tüketim verimliliğinin artırılması, sürücü endişesinin hafifletilmesine önemli ölçüde yardımcı olacak ve şarj altyapısının işletilmesi, planlanması ve yönetimi için temel bir çerçeve sağlayacaktır. Elektrikli araçların enerji tüketimi tahminindeki zorlukların üstesinden gelmek için veriler Japonya'nın Aichi Eyaletinde toplanmıştır. Çalışmada, elektrikli araçların enerji tüketiminin tahmini için geleneksel makine öğrenimi modelleri, Multi Output, Gradient Boosting, XGBoost ve Random Forest kullanılmıştır. Tahmin modellerinin performansını değerlendirmek için belirleme katsayısı (R^2), kök ortalama kare hatası (RMSE) ve ortalama mutlak hata (MAE) değerlendirme ölçütleri kullanılmıştır. Tahmin sonuçları, Gradient Boosting ve Multi Output birleşimi ile oluşturulan regresyon modeli iyi performans gösterdiğini ortaya koymaktadır. Daha yüksek R^2 değerlerine, daha düşük MAE ve RMSE değerlerine sahip Gradient Boosting ve Multi Output tabanlı modellerin daha doğru olduğu kanıtlanmıştır. Farklı girdi değişkenlerinin elektrikli araçların enerji tüketimi tahmini üzerindeki etkisini ve göreceli etkisini göstermek için ayrıntılı bir önemli özellik analizi gerçekleştirilmiştir. Sonuçlar, gelişmiş bir makine öğrenmesi modelinin elektrikli araçların enerji tüketiminin tahmin performansını artırabileceğini göstermektedir.

References

  • Abdelaty, H., Al-Obaidi, A. & Mohamed, M. (2021). Machine learning prediction models for battery-electric bus energy consumption in transit. Transportation Research Part D: Transport and Environment, 96, 102868.
  • Alves, J., Baptista, P. C. & Gonçalves, G. A. (2016). Indirect methodologies to estimate energy use in vehicles: Application to battery electric vehicles. Energy Conversion and Management, 124, 116–129.
  • Amin, A., Altinoz, B. & Dogan, E. (2020). Analyzing the determinants of carbon emissions from transportation in European countries: The role of renewable energy and urbanization. Clean Technology and Environmental Policy, 22, 1725–1734.
  • Bi, J., Wang, Y. & Shao, S. (2018). Residual range estimation for battery electric vehicle based on radial basis function neural network. Measurement, 128, 197–203.
  • Bolovinou, A., Bakas, I. and Amditis, A.(2014). Online prediction of an electric vehicle remaining range based on regression analysis. In: 2014 IEEE International Electric Vehicle Conference (IEVC), pp. 1–8. IEEE.
  • Chen, Y., Wu, G. & Sun, R. (2020). A Review and Outlook of Energy Consumption Estimation Models for Electric Vehicles. ArXiv Preprint ArXiv200312873.
  • De Cauwer, C., Van Mierlo, J. & Coosemans, T. (2015). Energy consumption prediction for electric vehicles based on real-world data. Energies, 8, 8573–8593.
  • Felipe, J., Amarillo, J. C. & Naranjo, J. E. (2015). Energy consumption estimation in electric vehicles considering driving style. In: 2015 IEEE 18th International Conference on Intelligent Transportation Systems, pp. 101–106. IEEE.
  • Hertzke, P., Müller, N. & Schenk, S.(2018). The global electric-vehicle market is amped up and on the rise. McKinsey Center for Future Mobility, 1–8.
  • Hong, J., Park, S. & Chang, N. (2016). Accurate remaining range estimation for electric vehicles. In: 2016 21st Asia and South Pacific Design Automation Conference (ASP-DAC), pp. 781–786. IEEE.
  • Hughes, S., Moreno, S. & Yushimito, W. F. (2019). Evaluation of machine learning methodologies to predict stop delivery times from GPS data. Transportation Research Part C: Emerging Technologies, 109, 289–304.
  • Koubaa, R., Yoldas, Y. & Goren, S. (2020). Implementation of cost-benefit analysis of vehicle to grid coupled real Micro-Grid by considering battery energy wear: Practical study case. Energy & Environment, 0958305X20965158.
  • Li, P., Zhang, Y. & Zhang, Y. (2021). Prediction of electric bus energy consumption with stochastic speed profile generation modelling and data-driven method based on real-world big data. Applied Energy, 298, 117204.
  • Liu, K., Wang, J., Yamamoto, T., et al. (2016). Modelling the multilevel structure and mixed effects of the factors influencing the energy consumption of electric vehicles. Applied Energy, 183, 1351–1360.
  • Liu, K., Yamamoto, T. & Morikawa, T. (2017). Impact of road gradient on energy consumption of electric vehicles. Transportation Research Part D: Transport and Environment, 54, 74–81.
  • López, F. C., and Fernández, R. Á. (2020). Predictive model for energy consumption of battery electric vehicle with consideration of self-uncertainty route factors. Journal of Cleaner Production, 276, 124188.
  • Modi, S., Bhattacharya, J. & Basak, P. (2020). Estimation of energy consumption of electric vehicles using Deep Convolutional Neural Network to reduce driver’s range anxiety. ISA Transactions, 98, 454–470.
  • Paçal, İ. (2024). MaxCerVixT: A novel lightweight vision transformer-based Approach for precise cervical cancer detection, Knowledge-Based Systems, 289, 111482.
  • Paçal, İ. (2023). Göğüs röntgeni görüntülerinden otomatik covıd-19 teşhisi için görü transformatörüne dayalı bir yaklaşım. Iğdır Üniversitesi Fen Bilimleri Enstitüsü Dergisi 13, 2, 778–791.
  • Ullah, I., Liu, K. Yamamoto, T. Al Mamlook, R. E. & Jamal, A. (2022). A comparative performance of machine learning algorithm to predict electric vehicles energy consumption: A path towards sustainability. Energy & Environment, 33 (8), 1583-1612.
  • Vita, V., & Koumides, P. (2019). Electric vehicles and distribution networks: Analysis on vehicle to grid and renewable energy sources integration. In: 2019 11th Electrical Engineering Faculty Conference (BulEF), pp. 1–4. IEEE.
  • Wang, J., Liu, K. & Yamamoto, T. (2017). Improving electricity consumption estimation for electric vehicles based on sparse GPS observations. Energies, 10, 129.
  • Xu, G., Wang, S. & Li, J. (2020). Moving towards sustainable purchase behavior: Examining the determinants of consumers’ intentions to adopt electric vehicles. Environmental Science and Pollution Research, 1–12.

Comparing The Performance of Machine Learning Algorithms in Predicting Energy Consumption of Electric Vehicles

Year 2024, , 518 - 530, 01.06.2024
https://doi.org/10.21597/jist.1448216

Abstract

In order to maintain environmental sustainability, it is crucial to address the transportation sector's explosive growth and the emissions that accompany it. Therefore, understanding the derivation factor of transportation emissions is of utmost importance. In the face of increasing transport emissions, it is important to expand the role of electric vehicles (EVs). An eco-friendly economy and low-carbon economy are made possible by electric automobiles. Energy consumption models that can accurately and consistently forecast energy use are critical to the successful deployment of electric vehicles. Enhancing EVs' energy efficiency will help reduce driver anxiety a great deal and offer a foundation for organizing, operating, and managing the infrastructure for charging. To overcome the challenges in estimating the energy consumption of electric vehicles, data was collected in Aichi Prefecture, Japan. In the study, traditional machine learning models, Multi Output, Gradient Boosting, XGBoost and Random Forest were used to predict the energy consumption of electric vehicles. The prediction models were assessed using the coefficient of determination (R2), mean absolute error (MAE), and root mean square error (RMSE). The forecasting results reveal that the regression model with the combination of Gradient Boosting and Multi Output performs well. Gradient Boosting and Multi Output based models with higher R2 values and lower MAE and RMSE values are proved to be more accurate To demonstrate the effect and relative influence of various input variables on the energy consumption forecast of electric vehicles, a thorough important feature analysis is carried out. The findings demonstrate that a sophisticated machine learning model can enhance the accuracy of electric car energy consumption predictions.

References

  • Abdelaty, H., Al-Obaidi, A. & Mohamed, M. (2021). Machine learning prediction models for battery-electric bus energy consumption in transit. Transportation Research Part D: Transport and Environment, 96, 102868.
  • Alves, J., Baptista, P. C. & Gonçalves, G. A. (2016). Indirect methodologies to estimate energy use in vehicles: Application to battery electric vehicles. Energy Conversion and Management, 124, 116–129.
  • Amin, A., Altinoz, B. & Dogan, E. (2020). Analyzing the determinants of carbon emissions from transportation in European countries: The role of renewable energy and urbanization. Clean Technology and Environmental Policy, 22, 1725–1734.
  • Bi, J., Wang, Y. & Shao, S. (2018). Residual range estimation for battery electric vehicle based on radial basis function neural network. Measurement, 128, 197–203.
  • Bolovinou, A., Bakas, I. and Amditis, A.(2014). Online prediction of an electric vehicle remaining range based on regression analysis. In: 2014 IEEE International Electric Vehicle Conference (IEVC), pp. 1–8. IEEE.
  • Chen, Y., Wu, G. & Sun, R. (2020). A Review and Outlook of Energy Consumption Estimation Models for Electric Vehicles. ArXiv Preprint ArXiv200312873.
  • De Cauwer, C., Van Mierlo, J. & Coosemans, T. (2015). Energy consumption prediction for electric vehicles based on real-world data. Energies, 8, 8573–8593.
  • Felipe, J., Amarillo, J. C. & Naranjo, J. E. (2015). Energy consumption estimation in electric vehicles considering driving style. In: 2015 IEEE 18th International Conference on Intelligent Transportation Systems, pp. 101–106. IEEE.
  • Hertzke, P., Müller, N. & Schenk, S.(2018). The global electric-vehicle market is amped up and on the rise. McKinsey Center for Future Mobility, 1–8.
  • Hong, J., Park, S. & Chang, N. (2016). Accurate remaining range estimation for electric vehicles. In: 2016 21st Asia and South Pacific Design Automation Conference (ASP-DAC), pp. 781–786. IEEE.
  • Hughes, S., Moreno, S. & Yushimito, W. F. (2019). Evaluation of machine learning methodologies to predict stop delivery times from GPS data. Transportation Research Part C: Emerging Technologies, 109, 289–304.
  • Koubaa, R., Yoldas, Y. & Goren, S. (2020). Implementation of cost-benefit analysis of vehicle to grid coupled real Micro-Grid by considering battery energy wear: Practical study case. Energy & Environment, 0958305X20965158.
  • Li, P., Zhang, Y. & Zhang, Y. (2021). Prediction of electric bus energy consumption with stochastic speed profile generation modelling and data-driven method based on real-world big data. Applied Energy, 298, 117204.
  • Liu, K., Wang, J., Yamamoto, T., et al. (2016). Modelling the multilevel structure and mixed effects of the factors influencing the energy consumption of electric vehicles. Applied Energy, 183, 1351–1360.
  • Liu, K., Yamamoto, T. & Morikawa, T. (2017). Impact of road gradient on energy consumption of electric vehicles. Transportation Research Part D: Transport and Environment, 54, 74–81.
  • López, F. C., and Fernández, R. Á. (2020). Predictive model for energy consumption of battery electric vehicle with consideration of self-uncertainty route factors. Journal of Cleaner Production, 276, 124188.
  • Modi, S., Bhattacharya, J. & Basak, P. (2020). Estimation of energy consumption of electric vehicles using Deep Convolutional Neural Network to reduce driver’s range anxiety. ISA Transactions, 98, 454–470.
  • Paçal, İ. (2024). MaxCerVixT: A novel lightweight vision transformer-based Approach for precise cervical cancer detection, Knowledge-Based Systems, 289, 111482.
  • Paçal, İ. (2023). Göğüs röntgeni görüntülerinden otomatik covıd-19 teşhisi için görü transformatörüne dayalı bir yaklaşım. Iğdır Üniversitesi Fen Bilimleri Enstitüsü Dergisi 13, 2, 778–791.
  • Ullah, I., Liu, K. Yamamoto, T. Al Mamlook, R. E. & Jamal, A. (2022). A comparative performance of machine learning algorithm to predict electric vehicles energy consumption: A path towards sustainability. Energy & Environment, 33 (8), 1583-1612.
  • Vita, V., & Koumides, P. (2019). Electric vehicles and distribution networks: Analysis on vehicle to grid and renewable energy sources integration. In: 2019 11th Electrical Engineering Faculty Conference (BulEF), pp. 1–4. IEEE.
  • Wang, J., Liu, K. & Yamamoto, T. (2017). Improving electricity consumption estimation for electric vehicles based on sparse GPS observations. Energies, 10, 129.
  • Xu, G., Wang, S. & Li, J. (2020). Moving towards sustainable purchase behavior: Examining the determinants of consumers’ intentions to adopt electric vehicles. Environmental Science and Pollution Research, 1–12.
There are 23 citations in total.

Details

Primary Language Turkish
Subjects Computer Software, Electrical Engineering (Other)
Journal Section Bilgisayar Mühendisliği / Computer Engineering
Authors

Emrah Aslan 0000-0002-0181-3658

Early Pub Date May 28, 2024
Publication Date June 1, 2024
Submission Date March 6, 2024
Acceptance Date April 3, 2024
Published in Issue Year 2024

Cite

APA Aslan, E. (2024). Elektrikli Araçların Enerji Tüketimini Tahmin Etmede Makine Öğrenimi Algoritmalarının Performanslarının Karşılaştırılması. Journal of the Institute of Science and Technology, 14(2), 518-530. https://doi.org/10.21597/jist.1448216
AMA Aslan E. Elektrikli Araçların Enerji Tüketimini Tahmin Etmede Makine Öğrenimi Algoritmalarının Performanslarının Karşılaştırılması. Iğdır Üniv. Fen Bil Enst. Der. June 2024;14(2):518-530. doi:10.21597/jist.1448216
Chicago Aslan, Emrah. “Elektrikli Araçların Enerji Tüketimini Tahmin Etmede Makine Öğrenimi Algoritmalarının Performanslarının Karşılaştırılması”. Journal of the Institute of Science and Technology 14, no. 2 (June 2024): 518-30. https://doi.org/10.21597/jist.1448216.
EndNote Aslan E (June 1, 2024) Elektrikli Araçların Enerji Tüketimini Tahmin Etmede Makine Öğrenimi Algoritmalarının Performanslarının Karşılaştırılması. Journal of the Institute of Science and Technology 14 2 518–530.
IEEE E. Aslan, “Elektrikli Araçların Enerji Tüketimini Tahmin Etmede Makine Öğrenimi Algoritmalarının Performanslarının Karşılaştırılması”, Iğdır Üniv. Fen Bil Enst. Der., vol. 14, no. 2, pp. 518–530, 2024, doi: 10.21597/jist.1448216.
ISNAD Aslan, Emrah. “Elektrikli Araçların Enerji Tüketimini Tahmin Etmede Makine Öğrenimi Algoritmalarının Performanslarının Karşılaştırılması”. Journal of the Institute of Science and Technology 14/2 (June 2024), 518-530. https://doi.org/10.21597/jist.1448216.
JAMA Aslan E. Elektrikli Araçların Enerji Tüketimini Tahmin Etmede Makine Öğrenimi Algoritmalarının Performanslarının Karşılaştırılması. Iğdır Üniv. Fen Bil Enst. Der. 2024;14:518–530.
MLA Aslan, Emrah. “Elektrikli Araçların Enerji Tüketimini Tahmin Etmede Makine Öğrenimi Algoritmalarının Performanslarının Karşılaştırılması”. Journal of the Institute of Science and Technology, vol. 14, no. 2, 2024, pp. 518-30, doi:10.21597/jist.1448216.
Vancouver Aslan E. Elektrikli Araçların Enerji Tüketimini Tahmin Etmede Makine Öğrenimi Algoritmalarının Performanslarının Karşılaştırılması. Iğdır Üniv. Fen Bil Enst. Der. 2024;14(2):518-30.