ASHRAE AN-04-7-2-2004 Theoretical and Experimental Study of a New Absorption Refrigeration Cycle《新的吸收式制冷循环理论及实验研究》.pdf

1、AN-04-7-2 Theoretical and Experimental Study of a New Absorption Refrigeration Cycle Yongfang Zhong Student Member ASHRAE ABSTRACT A new auto-cascade absorption refrigeration cycle is proposed to obtain low refrigerating temperatures in order to widen the industry application ofabsorption refrigerat

2、ion. The characteristics of the new cycle are analyzed and compared to those of the traditional single-eflect absorption refrigeration cycle theoretically. Preliminary experiments are conducted and provide valuable experience for further research. INTRODUCTION With the rapidly rising cost of energy,

3、 low-temperature- level heat that was formerly rejected to the atmosphere in chemical and food plants is now often used to operate absorp- tion systems for refrigeration. Heat-driven absorption refng- eration, instead of work-driven compression refrigeration, greatly reduces expenses by utilizing wa

4、ste heat in applica- tions in many fields, such as chemistry and the food industry. In order to promote the use of absorption systems, a lot of work has been done to improve its efficiency (Srikhirin et a. 2002). However, the refrigerating temperature of the absorption cycle restricts its industrial

5、 application. Absorption machines commercially available today employed below 0C are ammo- nialwater systems, with ammonia as their refrigerant and water as their absorbent. The strong toxicity and highly irri- tating odor of ammonia are serious obstacles to its widespread use. If absorption systems

6、 can be widely used to obtain low temperatures, a portion of the heat rejected by a power plant, for example, can be utilized to freeze foods. The overall energy efficiency can be improved significantly. In this paper, a new auto-cascade absorption refrigeration cycle using a zeotropic mixture as a

7、refrigerant and DMF as an Guangming Chen absorbent is proposed to achieve low temperatures. The selec- tion of working fluids is based on our previous experiments and a literature review of absorption refrigerants. (N,N-dime- thylformamide, which is an organic solvent produced in large quantities th

8、roughout the world, is used in the chemical indus- try as a solvent, an intermediate, and an additive. It is a color- less liquid with a faint amine odor. It is completely miscible with water and most organic solvents and has a relatively low vapor pressure.) A diagram of the new auto-cascade absorp

9、tion refngera- tion system is shown in Figure 1. This cycle is derived from a compression refrigeration cycle applied to a liquefaction system for natural gas, where high-boiling-point gases are first liquefied and then used to condense low-boiling-point gases in cascade heat exchangers (Kleemenko 1

10、959; Little 1982). In Figure 1 Diagram of the new auto-cascade absorption refrigeration system. Yongfang Zhong is a graduate student in the Department of Mechanical Engineering, University of Illinois at Urbana-Champaign, Urbana, Ill. Guangming Chen is a professor at the Refrigeration and Cryogenic

11、Lab, Zhejiang University, China. 508 02004 ASHRAE. our new cycle, the refrigerant loop takes the mixed refrigerant vapor from the generator and directs it to the condenser where the high-boiling-point refrigerant is liquefied and the low- boiling-point refrigerant is subcooled by rejecting heat to a

12、 sink. The two-phase mixture is separated into vapor (which is rich in low-boiling-point refngerant and flows out from the top) and liquid (which is rich in high-boiling-point refrigerant and flows out from the bottom) at the separator. In this paper we call the vapor from the separator “Fl” and the

13、 liquid out of the separator “F2.” Then, F2 flows through a throttling valve into a cascade heat exchanger, where F2 evaporates to condense F1. After FI is liquefied in the heat exchanger, it passes through a cooling regenerator, then flows into an evap- orator via a throttling valve and evaporates

14、to provide a refrig- erating effect. Finally, F 1 from the evaporator and F2 from the cascade heat exchanger enter the absorber as a mixture. The solution circuit is the same as a single-effect absorption cycle, in which one condenser, one evaporator, one generator, and one absorber are used. In our

15、 new absorption cycle, the refrig- erant in the evaporator is subcooled both in the condenser and in the cascade heat exchanger before it evaporates to provide refrigerating effect. Therefore, it can provide lower refriger- ating temperatures than the traditional cycle, with one condenser under the

16、same operating conditions. Because a part of the refrigerants rejects heat to the other part of the refrigerants in a cascade heat exchanger and then is condensed, we call this cycle the “auto-cascade” absorption refrigeration cycle. The performance characteristics such as refrigerant composition an

17、d condensing temperature for this new cycle are investigated in this paper. In addition, the characteristics of the traditional single-effect absorption refrigeration cycle are analyzed under the same operating conditions. Some prelim- inary experiments are conducted to assess the feasibility of thi

18、s new cycle for low refrigerating temperature application. THEORETICAL CALCULATION The operating conditions discussed here are as follows: Tl = Tg= 140C, TIO= T,=-47”C, Tk= 35C (condensing temper- ature), T3=30”C, Tl,=30”C,P3,4= 100kPa,AT1=3”C(inthe cascade heat exchanger and the regenerator), AT2 =

19、 8C (in the heat exchanger in the solution circuit), and environmentally friendly R-23 and R-l34a are used in the calculation as mixed refrigerants. The refrigerating effect (e,) and COP (ratio of Q, to the heat required in the generator to accomplish this effect, Qg) under the various operating con

20、ditions are investigated with other conditions fixed. The characteristics of the tradi- tional single-effect cycle are also calculated for comparison for the same conditions. The analysis here is ideal thermody- namic calculation; energy change of the flow equals mass times change of enthalpy; the f

21、luid properties at each state point are calculated by thermal equations of state and two- phase equilibrium. All the calculation for the cycle and heat exchangers is based on mass and energy balance. Characteristics Analysis of the New Cycle COP and Q, vs. refrigerant composition. If the molar ratio

22、 of R-23 (molar ratio of R-23 is the ratio of the amount of R-23 in mole to the total amount of refriger- ants in mole) in the mixture is too low or too high, there will be little liquid of R-23 in the evaporator, and the new absorp- tion system cannot work normally. Therefore, the molar ratio of R-

23、23 discussed here ranges from 0.2 to 0.7. The refrigerating effect (Q,) and COP increase to a maxi- mum, then decrease as the molar ratio of R-23 increases (shown in Figure 2). The flow rate at point 5 (shown in Figure 1) increases as the molar ratio of R-23 increases; thus, the amount of heat (e5,)

24、 that is released from F1 in the cascade heat exchanger in order to lower the temperature of point 9 to (T7 + AT,) also increases. Simultaneously, the flow rate at point 6 decreases; thus, the amount of heat that is neces- sary to increase the temperature of point 8 to (T, - ATl) also decreases. The

25、refore, the heat load in the cascade heat exchanger is the minimum of QSg and Q78. (In this paper, we call the amount of heat being released from F1 to decrease the temperature of point 9 to (T7 + ATl) as the heat capacity of F1; we call the amount of heat needed for F2 to increase the temperature o

26、f point 8 to T5 - ATl as the heat capacity of F2.) When it reaches its maximum load as the refrigerant composition varies (i.e., the molar ratio of R-23 is approximately equal to 0.35), Q, reaches its maximum. As a result, COP reaches its maximum. Therefore, Q, and COP reach their maximum when the h

27、eat capacities of F1 and F2 in the cascade heat exchanger are approximately equal, as the molarratio ofR-23 varies from0.2 to 0.7. A molar ratio of0.35 for R-23 is used in the following analysis of characteristics for the new cycle. O. 14 o. 12 o. 1 O. 08 O. 06 O. 04 o. 02 O CL O V o. 1 o. 3 o. 5 o.

28、 7 o. 9 R23 (Mol%) Figure 2 COP and Q, vs. refrigerant composition. ASHRAE Transactions: Symposia 509 O. 14 o. 12 o. 1 o. O8 O. 06 c, U O. 04 o. 02 O 32 34 36 38 40 42 44 46 Tk(C) 10 9 8 o. 2 O. 18 O. 16 O. 14 o. 12 a 8 o. 1 O. 08 O. 06 O. 04 o. o2 O 230 250 270 290 310 330 Tg(C) 10 9 8 Figure 3 COP

29、 and Q, vs. condensing temperature (R-23 = 35mo1%). 35mo1%). Figure 4 COP and Qe vs. generating temperature (R-23 = o. 3 8700 COP and Q, vs. condensing temperature. The amount of heat required to be transferred into the generator (Q,) decreases while the condensing temperature increases. The flow ra

30、tes of F 1 and F2 change in the cascade heat exchanger as the condensing temperature varies, resulting in a change of the heat capacities of F1 and F2 (eS9 and Q78). At the beginning, Q, increases as the condensing temperature increases. It achieves a maximum as the heat capacities of F1 and F2 matc

31、h well in the cascade heat exchanger, i.e., when the amount of heat being released from F1 to decrease the temper- ature of point 9 to (T7 + AT,) equals the amount of heat needed for F2 to increase the temperature of point 8 to (T, - AT,). After that, Q, begins to decrease. Simultaneously, Qg decrea

32、ses more rapidly than Q,; thus, both Q, and COP achieve the maximum when the heat capacities of F1 and F2 are approximately equal in the cascade heat exchanger, as shown in Figure 3. This situation differs greatly from the tradi- tional absorption system where the COP and Q, always decrease as the c

33、ondensing temperature increases. COP and Q, vs. generating temperature. h c 0.05 O - 8350 -32 -36 -40 -44 -48 -52 TlO(C) Figure 5 COP and Q, vs. evaporating temperature (R-23 = 35mo/%). When the generating temperature increases, the differ- ence of the solution concentration (refrigerant concentrati

34、on in DMF between the solutions from the absorber and out of the generator) becomes larger, resulting in less heat being required in the generator (eg). However, the refrigerating effect (Q,) is unlikely to be affected by the generating temper- ature. Hence, Q, remains invariable, so COP is improved

35、 after the generating temperature increases, as shown in Figure 4. when the refrigerating temperature decreases and, thus, leads to a larger Q,. The pressure of F2 at point 7 in Figure 1 decreases with the refrigerating temperature. Consequently, the pressure decreases the temperature and enthalpy o

36、f F1 at point 9, resulting in a larger Q,. With a more rapid increase of Qg than that of Q, COP decreases slightly as the refrigerating temperature decreases. COP and Q, vs. evaporating temperature. The impact of evaporating temperature on performance characteristics for the new cycle is shown in Fi

37、gure 5. The concentration difference of DMF solution becomes smaller All of the above calculations show whether or not the heat capacities of the two flows, F1 and F2 (Q, and Q7 this evaporation allows it to achieve much lower refrigerating temperatures under the same operating conditions. The COP a

38、nd the refrigerating temperature of the new cycle are affected by the molar ratio of the refrigerant mixture. For example, the minimum refrigerating temperature of the new cycle with R-23 = 35mol% is lower than that of a cycle with R-23 = 75mol%. The COP of the new cycle with R-23 = 35mol% is much l

39、arger than that of the cycle with R-23 = 75mol% at the same refrigerating temperature. The new cycle will use the mixture of R-23 (35mol%) and R-134a (65mol%) as its refrigerant in the following discussion. COP vs. refrigerating temperature using R- 134a in tradi- tional cycle. When the traditional

40、cycle uses pure R- 134a as its refrig- erant and the new cycle uses R-23 (35mol%) and R-134a (65mol%) (see above), the relationship between the COP and o. 35 o. 3 O. 25 o. 2 O. 15 o. 1 O. 05 O p. O U -40 -30 -20 -10 O 10 20 T10 (C) Figure 6 COP vs. refrigerating temperature using R-23 + R-134a (R-23

41、 = 35mol%). the refrigerating temperature is analyzed as shown in Figure 8 (COPR134a represents the traditional cycle). The new cycle can reach much lower temperatures than the traditional cycle under the same operating conditions. However, the COP of the new cycle may be much less than that of the

42、traditional cycle, especially when the evaporating temperature is above 0C. Most likely, this performance is due to the larger latent heat of R-134a than that of R-23. It may also be caused by inferior matching of the heat capacities of F1 and F2 in the cascade heat exchanger as refrigeration temper

43、ature increases. 0.2 I I O. 18 O. 16 O. 14 o. 12 0 o. 1 U O. 08 O. 06 O. 04 o. 02 O -20 -10 O 10 T10 (C Figure 7 COP vs. refrigerating temperature using R-23 + R-134a (R-23 = 75mo1%). O. 6 o. 5 o. 4 u 0.3 o. 2 o. 1 O a -35 -25 -15 5 15 TIO (ci5 Figure 8 COP vs. refrigerating temperature using R-134a

44、 in traditional cycle. ASH RAE Transactions: Symposia 51 1 Heat Exchangtr From regmaator and r-evaporator Figure 9 Diagram of the solution circuit in experiments. COP vs. refrigerating temperature using R-23 in the tra- ditional cycle. When pure R-23 is used in the traditional cycle, it leads to a v

45、ery poor refrigerating performance-its COP gets zero when its refrigerating temperature is around O“C, and its COP is around 0.025 when its refrigerating temperature is around 5C. A possible reason is the difficulty for pure R-23 to condense under the designed operating conditions. Obviously, pure R

46、-23 is not suitable as the refrigerant in such an absorp- tion system. The new absorption cycle theoretically shows a better potential to accomplish lower refrigeration temperature than the traditional one according to the above analysis. Moreover, a multiple-stage and multiple-effect solution circu

47、it can be applied to this new cycle to rther improve its efficiency. EXPERIMENTAL STUDY Apparatus and Method A diagram of the solution circuit used in the experiments is shown in Figure 9. When heat is applied to the solution in the generator, vapor evaporates and flows into the condenser. The remai

48、ning liquid solution leaves the generator and flows back to the absorber through two heat exchangers and a throt- tling valve V2. This solution absorbs the refrigerant from the regenerator and the cascade heat exchanger, respectively, and becomes a relatively concentrated refrigerant solution. The c

49、oncentrated solution leaves the absorber to a tank by gravity and is then pumped into the generator, The tank, pump, and the two valves, V1 and V3, constitute a bypass system to control the flow rate of the solution to the generator. Two level indi- cators are used-one in the generator and the other in the absorber-to show the solution level and help to control the system in experiments. To condensa From pump 1 ptifmtcd piate Ztlectic heater 3 he1 indicator 2 c, Toabsorber Figure 10 Diagram of the generator. strow solution out Figure Il Diagram of the absorbez The designs of t