Composite desirability function-based empirical modeling for packed tower design in physical ammonia absorption


Yetilmezsoy K.

ASIA-PACIFIC JOURNAL OF CHEMICAL ENGINEERING, cilt.7, ss.795-813, 2012 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 7
  • Basım Tarihi: 2012
  • Doi Numarası: 10.1002/apj.635
  • Dergi Adı: ASIA-PACIFIC JOURNAL OF CHEMICAL ENGINEERING
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.795-813
  • Anahtar Kelimeler: ammonia absorption, packed tower, column diameter, packing height, empirical modeling, NEURAL-NETWORK, CHEMICAL ABSORPTION, PILOT-SCALE, AIR, PREDICTION, PERFORMANCE, EFFICIENCY, OPTIMIZATION, REGRESSION, COLUMNS
  • Yıldız Teknik Üniversitesi Adresli: Evet

Özet

This paper describes two new empiricial models based on the composite desirability function methodology for the prediction of packed column diameter and packing height in solving problems of physical ammonia absorption. The computational analysis including a total of 131 absorption scenarios were conducted for six fundamental operating parameters such as inlet gas flow rate (5005000?m3/h), solute concentration in the inlet gas (510%), desired removal efficiency (8798%), operating temperature (1060 degrees C), percentage of the minimal liquid-to-gas flow rates ratio (2452%) and percent of flooding velocity (3080%). The ceramic Raschig rings were selected as one of the most common packing types for the present analysis. The column diameter to packing size ratio were considered to be greater than 30 for a total of 500 packing sizes (0.015?inches in diameter) included in the numerical cycle. New non-linear regressions as the functions of packing factor, packing size and various transfer unit constants (alpha, beta, gamma, ?eta and phi) were implemented in the prediction scheme of the proposed empirical models. The performance of the models was evaluated with various descriptive statistical indicators. The testing results showed that all predictions were proven to be satisfactory with very high determination coefficients of about R2?=?0.997 and R2?=?0.963 for the packed column diameter and required packing height, respectively. (c) 2011 Curtin University of Technology and John Wiley & Sons, Ltd.