Yazar "Ozalp, Coskun" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • [ X ]
    Öğe
    Experimental and Modeling Study of Peanut Drying in a Solar Dryer with a Novel Type of a Drying Chamber
    (Taylor & Francis Inc, 2022) Hurdogan, Ertac; Cerci, Kamil Neyfel; Saydam, Dogan Burak; Ozalp, Coskun
    The aim of the present study is to investigate the experimental performance of a solar dryer by using a novel type of a drying chamber for increasing the drying performance and ensuring homogeneous drying. The drying system (SDS) consisted of a new type of a drying chamber and a solar air collector, which were designed, made, and tested in Osmaniye, Turkey. The analyses of Computational Fluid Dynamics were carried out to demonstrate the advantage of the designed drying chamber over the tunnel type drying chamber, which is commonly used in peanut drying. The drying experiments were conducted using in-shell peanut to observe the performance of the system. The drying procedure was performed in the form of open sun drying (OSD), and the results of drying kinetics such as moisture ratio (MR), drying rate (DR), and heat transfer coefficient (h(c)) obtained from both drying methods were compared. The MR and DR values obtained during the drying experiments were estimated using different models such as mathematical, multiple linear regression (MLR), decision tree (DT), and the support vector machine (SVM). It was observed from the numerical results that the products can dry more homogenously and effectively with the newly designed drying chamber compared to the conventional tunnel type drying chamber. The results of the drying experiments showed that the products dried earlier and more regularly through the SDS compared to the OSD. The DR and h(c) values were found 0.0051E-01 (g(w)/g(dm))/min and 1.5727 W/m(2)degrees C for SDS and 0.0039E-01 (g(w)/g(dm))/min and 1.4664 W/m(2)degrees C for OSD, respectively. The models that best estimated the experimentally obtained MR and DR values for peanuts dried with the SDS proved to be the Random Tree Model (R-2 = 0.9972) and Quintic Model (R-2 = 0.8551), respectively.
  • [ X ]
    Öğe
    Experimental investigation and artificial neural networks (ANNs) based prediction of thermal performance of solar air heaters with different surface geometry
    (Pergamon-Elsevier Science Ltd, 2024) Cerci, Kamil Neyfel; Saydam, Dogan Burak; Hurdogan, Ertac; Ozalp, Coskun
    Solar air heaters (SAHs) are one of the most utilized tools to obtain heat energy from the sun. A variety of SAH models exist with different geometries used for improving the heat transfer between the absorber plate and the air in the SAHs. In today ' s world, many researchers are focusing on designs that occupy the same dimensions but can generate more useful energy. In this study, two SAH with the same external volume (same base area and height), but different absorber surface geometries: a V -corrugated type (V -Type) and an internal baffle -installed type (B -Type), were designed, manufactured and experimentally tested in the climatic conditions of Osmaniye, T & uuml;rkiye. B -Type SAH, unlike the literature, is constructed with 11 inner baffles to allow the air to circulate more within the collector, aiming to achieve higher temperatures. The performances of the SAHs were compared by means of energy, exergy and enviro-economic (3E) analyses in the experiments lasting four consecutive days without interruption. The results show that the average amount of useful heat, energy efficiency ( eta I ) and exergy efficiency ( eta II ) of the V -Type SAH was 35.71%, 38.00% and 80.11% higher, respectively, than those obtained with the B -Type SAH, and also that the V -Type SAH was more environmentally friendly than the B -Type SAH. Additionally, in this study, common ANN models predicting the performance parameters of both SAHs were developed, constituting the another novelty of the research. Among the common ANN models developed for the outlet temperature ( T o ) , besides the eta I and eta II parameters, the best results were obtained with ANN 15, ANN 13 and ANN 18, respectively. Therefore, it is possible to use these developed models safely.