Development of an artificial intelligence-based cooling load map for buses in Türkiye

Ünal Ş., BİLGİLİ M., BÜYÜKALACA O., Akgün H.

Thermal Science and Engineering Progress, cilt.61, 2025 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 61
  • Basım Tarihi: 2025
  • Doi Numarası: 10.1016/j.tsep.2025.103516
  • Dergi Adı: Thermal Science and Engineering Progress
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Anahtar Kelimeler: Artificial neural network, Bus, Cooling load, Machine learning, Radiant time series method, Vehicle
  • Çukurova Üniversitesi Adresli: Evet

Özet

Accurate calculation of air-conditioning cooling load is a prerequisite for optimal design of vehicle air-conditioning systems. This study aims to develop a monthly average bus cooling load map for Türkiye using an artificial neural network (ANN) approach. For this purpose, the Radiant Time Series (RTS) method recommended by ASHRAE was used as the cooling load calculation method. The cooling load values obtained for all 81 cities of Türkiye then utilized to train the ANN model, and a cooling load map was created that allows for the determination of the cooling load value for buses across the country. A comparison of the results from the RTS and ANN methods revealed high similarity, with Adana showing the highest cooling load and Ardahan the lowest in both models. The difference between maximum cooling load values from RTS and ANN ranged from 13.4% to 8.8% across test cities. Ultimately, the ANN approach offers a rapid and reliable means of calculating bus cooling loads, allowing for predictive adjustments to air-conditioning systems and potential energy savings.