## Genetik Algoritma Tabanlı Adsorpsiyon İzoterm Parametre Kestirimi

“International Congress on Human-Computer Interaction, Optimization and Robotic Applications”, Nevşehir, Türkiye, 5 - 07 July 2019, ss.1

• Basıldığı Şehir: Nevşehir
• Basıldığı Ülke: Türkiye
• Sayfa Sayısı: ss.1

#### Özet

Çalışılan 15 izoterm modeli için izoterm katsayıları ve izoterm eğrileri elde edilmiştir. Üç farklı sıcaklıktaki deneysel veriler için ayrı ayrı RMSE (Root Mean Square Error) hata değerleri hesaplanmıştır.  En küçük hata değerleri 25, 35, 45 oC’ler için sırasıyla 4.6×10-4  (Halsey izotermi), 3.19×10-4 (Halsey izotermi) ve 1.32×10-2 (Baudu izotermi) olarak bulunmuştur.  En büyük RMSE değerleri ise 1.77×10-1; 2.10×10-1 ve 5.23×10-1 olarak hesaplanmıştır. Bu değerler sırasıyla 25 oC için Fritz- Schlunder (IV), 35 oC için Langmuir ve 45 oC için Freundlich izotermleri içindir.

The process of removing substances dissolved in a solution with a substance called adsorbent is called adsorption. The amount of material that is desired to be removed from the solution during this process generally varies depending on the nature of the adsorbent used, the physical size and also the initial density of the material to be adsorbed in the solution. The amount of material adsorbed by the adsorbent increases at the beginning of the adsorption process and remains constant after a long enough time. In the literature, there are dozens of methods called isotherms that model this process in terms of the amount of substance removed and the substance in solution. Each model contains a different number of parameters. Isotherm parameters are estimated by using different experimental methods. In this study, 15 different isotherm models were applied to experimental adsorption data obtained from literature and the isotherm parameters were estimated by using genetic algorithm based optimization also model errors were calculated.

Isotherm constants and isotherm curves were obtained for 15 different isotherm models and then RMSE (Root Mean Square Error) error values ??were calculated for experimental data for three different temperatures.  The fact that the RMSE value is close to zero is a sign of the model and its compatibility with the experimental data.The smallest error values ??were 4.6×10-4  (Halsey isotherm), 3.19×10-4 (Halsey isotherm) and 1.32×10-2  (Baudu isotherm) for 25, 35, 45°C, respectively. The biggest RMSE values ??are 1.77×10-1; 2.10×10-1  and 5.23×10-1. These values ??are for Fritz- Schlunder (IV) for 25°C, Langmuir isotherms for 35°C and Freundlich isotherm 45°C respectively.