Estimation of the Effect of Oblique Positioned Obstacle Placement on Thermal Performance of a Horizontal Mantle Hot Water Tank with Machine Learning

Durmuşoğlu A., Turgut B., Tekin Y., Turgut B.

APPLIED SCIENCES, cilt.15, sa.48, ss.1-24, 2025 (SCI-Expanded)

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 15 Sayı: 48
  • Basım Tarihi: 2025
  • Doi Numarası: 10.3390/app15010048
  • Dergi Adı: APPLIED SCIENCES
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Aerospace Database, Agricultural & Environmental Science Database, Applied Science & Technology Source, Communication Abstracts, INSPEC, Metadex, Directory of Open Access Journals, Civil Engineering Abstracts
  • Sayfa Sayıları: ss.1-24
  • Hakkari Üniversitesi Adresli: Evet

Özet

Due to the growing popularity of vacuum tube solar collectors and their more esthetically pleasing look, horizontal hot water tanks are increasingly being used in solar hot water systems. In order to improve the thermal performance of a horizontal mantled hot water tank, this work numerically examines the impact of positioning inclination barriers parallel or coincident to one another at varying angles. The main input provided the velocity V = 0.036, 0.073, 0.11, and 0.147 m/s, and analysis were performed for each speed. The study concluded that V = 0.073 m/s was the ideal mains input velocity for each scenario and that raising the speed typically resulted in a lower mains outlet temperature. According to the study’s findings, the tank design with the first obstacle 150 mm away and the two obstacles 100 mm apart achieves the best efficiency. The residential water temperature in this model is 312 K, while the storage water temperature is 309.5 K. In this study, a feed-forward artificial neural network (ANN) model based predictor was designed to estimate the mantle outlet and main outlet temperatures and the temperature of the stored water. Analyses were performed for different network inlet velocities and obstacle combinations, and ANN showed superior performance in estimating temperature parameters.