Experimental analysis of CPV/T solar dryer with nano-enhanced PCM and prediction of drying parameters using ANN and SVM algorithms

Karaağaç M. O., Ergün A., Ağbulut Ü., Gürel A. E., Ceylan İ.

Solar Energy, cilt.218, ss.57-67, 2021 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 218
  • Basım Tarihi: 2021
  • Doi Numarası: 10.1016/j.solener.2021.02.028
  • Dergi Adı: Solar Energy
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, PASCAL, Aerospace Database, Applied Science & Technology Source, Communication Abstracts, Compendex, Computer & Applied Sciences, Environment Index, Geobase, INSPEC, Metadex, Civil Engineering Abstracts
  • Sayfa Sayıları: ss.57-67
  • Anahtar Kelimeler: Nano-enhanced PCM, Nanoparticle, Paraffin wax, Prediction algorihtms, Solar Drying, Thermal energy storage
  • Yıldız Teknik Üniversitesi Adresli: Hayır

Özet

In this paper, a concentrated photovoltaic-thermal solar dryer (CPV/TSD) using nano-enhanced PCM (Al2O3-Paraffin wax) is experimentally studied. A comprehensive thermodynamic analysis of the system according to the first and second laws is discussed. Besides, the drying parameters (moisture content and moisture ratio) are predicted using the two machine learning algorithms (ANN and SVM) and compared the prediction success with four evaluation metrics (R2, rRMSE, MBE, and rMAE). The overall thermal energy efficiency and exergy efficiency of the CPV/TSD system are found to be 20% and 8%, respectively. Although solar radiation to the environment has decreased a lot, it has been found that the thermal energy transferred to the nano-enhanced PCM prevents the decrease in greenhouse temperature for the first 100 min. In the system, mushrooms are dried from the initial moisture content of 17.45 g water/g dry matter to the final moisture content of 0.0515 g water/g dry matter. Then the drying rate value for CPV/TSD system is calculated to be 0.436 g matter/g dry matter.min. On the other hand, even if both ANN and SVM algorithms have exhibited very satisfying results, ANN is coming to the fore in the prediction of the drying parameters considering all evaluation metrics together.