TY - JOUR
TI - An adaptive neuro-fuzzy model of a re-heat two-stage adsorption chiller
AU - Krzywanski Jaroslaw
AU - Grabowska Karolina
AU - Sosnowski Marcin
AU - Zylka Anna
AU - Sztekler Karol
AU - Kalawa Wojciech
AU - Wojcik Tadeusz
AU - Nowak Wojciech
JN - Thermal Science
PY - 2019
VL - 23
IS - 14
SP - 1053
EP - 1063
PT - Article
AB - Since the adsorption chillers do not use primary energy as driving source the  possibility to employ low temperature waste heat sources in cooling energy  production receives nowadays much attention of the industry and science  community. However, the performance of the thermally driven adsorption  systems is lower than that of other heat driven heating/cooling systems. Low  coefficients of performance are one of the main disadvantages of adsorption  coolers. It is the result of a poor heat transfer coefficient between the  bed and the immersed heating surfaces of a built-in heat exchanger system. The purpose of this work is to study the effect of thermal conductance  values of sorption elements and evaporator as well as other design  parameters on the performance of a re-heat two-stage adsorption chiller.  One of the main energy efficiency factors in cooling production, i. e.  cooling capacity for wide-range of both design and operating parameters is  analyzed in the paper. Moreover, the work introduces artificial  intelligence approach for the optimization study of the adsorption cooler.  The ANFIS was employed in the work. The increase in both the bed and  evaporator conductance provides better performance of the considered  innovative adsorption chiller. The highest obtained value of cooling  capacity is 21.7 kW and it can be achieved for the following design and  operational parameters of the considered re-heat two-stage adsorption  chiller: Msorb = 40 kg, t = 1300 s, T = 80ºC, Csorb/Cmet = 50, hAsorb =  4000 W/K, hAevap = 4000 W/K.