Optimization of a lithium bromide–water solar absorption cooling system with evacuated tube collectors using the genetic algorithm
Introduction
Lithium bromide–water absorption cooling systems are economical due to their excellent performance and low costs. In these systems thermal energy is used to produce cooling and hence solar energy, waste heat and other forms of low grade energy can be applied to operate the system. Since the maximum cooling load occurs when the highest solar radiation is available, absorption systems have become more interesting.
Many researchers have studied the solar cooling and air-conditioning systems so far. A theoretical microcomputer model was presented by Tsilingiris [1] to evaluate the operational behavior of a simple solar LiBr–H2O absorption cooling system with 7 kW cooling capacity for small residential applications in Greece. The storage tank in the system was well mixed and heat losses in the interconnecting pipelines were ignored. The cooling system was simulated based on the manufacture's data of a typical LiBr absorption chiller.
A solar absorption system for a typical house in Beirut was simulated by Ghaddar et al. [2]. The absorption system was analyzed based on a thermodynamic model and the storage tank was supposed to be well mixed. The cooling capacity of the evaporator was variable. It was concluded that for each ton of refrigeration, a minimum flat plate collector of area 23.3 m2 with an optimal water storage tank capacity ranging between 1000 and 1500 l were required for the system to operate exclusively on solar energy for about 7 h a day. The performance of a 1.5 t solar cooling unit was evaluated by Hammad and Zurigat [3]. The system consisted of a 14 m2 flat-plate solar collector and five shell and tube heat exchangers. The unit was tested in April and May in Jordan and the maximum actual coefficient of performance obtained was 0.55. A system comprised of a solar collector, a storage tank, a boiler and a LiBr–H2O absorption system to provide the yearly cooling load of 78,235 MJ with a maximum hourly load of 40 MJ and a yearly heating load of 12,528 MJ with a maximum hourly load of 51.6 MJ for a typical house with 196 m2 floor area was simulated by Florides et al. [4]. The system was modeled in the TRANSYS environment and the program consists of many subroutines to model the subsystem components. The results show that it is not economical to replace any amount of load with heat collected from solar collector systems, but a dramatic increase in recent fuel price will change this result. A single-effect absorption chiller was simulated and the sensitivity analysis was conducted by Mehrabian and Shahbeik [5]. The results deduced from the computer program were used to study the effect of design parameters on cycle performance. It was concluded that increasing the evaporator and generator temperatures or decreasing the condenser and desorber temperatures can improve the second-law efficiency of the cycle. A solar LiBr–H2O absorption cooling system with an evacuated tube collector was simulated and optimized in TRANSYS environment for Malaysia's climate by Assilzadeh et al. [6]. It was proposed that a 0.8 m3 hot water storage tank and a collector of area 35 m2 at 20° slope angle for a 3.5 kW cooling load was required to achieve continuous operation and increase the reliability of the system. A solar single-effect lithium bromide–water absorption cooling system was modeled in Ahvaz by Mazloumi et al. [7]. The solar energy was absorbed by a horizontal N–S parabolic trough collector and stored in an insulated thermal storage tank. The results show that the collector mass flow rate had a negligible effect on the minimum required collector area, but it had a significant effect on the optimum capacity of the storage tank.
A full simulation model for absorption cooling systems, combined with a stratified storage tank, steady-state or dynamic collector model, and hourly building loads was achieved by Eicker and Pietruschka [8]. In order to validate the above-mentioned model, they used experimental data from various solar cooling plants. Since the absorption chillers could be run at reduced generator temperatures under partial load conditions, the control strategy had an important effect on the solar thermal performance and design. To study the influence of the special load time series for a given cooling power, various cooling load files with a dominance on either external or internal loads for different building locations and orientations were created. It was concluded that largely different collector areas and storage volumes, depending on the characteristics of the building load file and the chosen system technology and control strategy were needed to achieve a given solar fraction of the total heat demand. The results also showed that doubling the collector mass flow decreased the collector area resulting in reducing solar thermal system costs by 30%. Tsoutsos et al. [9] studied the performance and economic evaluation of a solar heating and cooling system of a hospital in Crete using the transient simulation program (TRNSYS). Simulating a complete system consisted of a solar collector, a storage tank, a backup heat source, a water cooling tower, and a LiBr–H2O absorption chiller were the main objective of this research. A fraction of the total cooling and heating energy demands of a hospital in Crete throughout the year was provided by this system. Some parameters namely Collector area, collector slope angle, volume of hot water storage tank, nominal power of absorption chiller, cooling tower and backup heat source were considered as the optimization parameters. It was deduced that the investment cost was quite high but the highest environmental benefits, the lower payback time, and the highest total annual savings compensated this. Lizarte et al. [10] investigated an innovative solar-driven, directly air-cooled, single-effect, 4.5 kW LiBr–H2O absorption chiller prototype. Air-conditioning a 40-m2 room located in Madrid was the goal of this project. Its solar facilities were a vacuum flat-plate collector field, with a total aperture area of 42.2 m2, a 25-kW external plate heat exchanger and a 1.5-m3 storage tank. The results comprised of the information including solar cooling facility temperatures and thermal power, solar fraction, COP and SCOP (solar coefficient of performance) values on a day when the outdoor dry bulb temperature ranged from 30 to 37.7 °C. It was observed that despite the high temperatures (109 °C) of the fluid fed by the solar facility into the prototype generator, no LiBr–H2O crystallization was observed. Congradac and Kulic [11] accomplished the optimization of chillers operating using artificial neural networks and genetic algorithms. The process of making specific chiller models used for testing the results of application of the genetic algorithm in usage optimization was shown as well as the basic characteristics of artificial neural networks. The paper also provided the optimal criteria used to obtain optimization results. The results of the artificial intelligence methods in the chiller optimization were validated in addition to the simulation tools Simulink and EnergyPlus and through a series of experiments on a real office building. The experimental tests were achieved between March and September 2008. It was observed that the artificial intelligence methods could be implemented on all subsystems of a modern integrated BMS, such as the lighting subsystem, the subsystem for protection from solar radiation, the subsystem for error prediction, etc. Thermodynamic modeling of a double-effect LiBr–H2O absorption refrigeration cycle was simulated by Iranmanesh and Mehrabian [12]. Conductance of all components was evaluated based on the approach temperatures assumed as input parameters. The effect of input data on the cycle performance and the exergetic efficiency were also investigated. A dynamic analysis of a single-effect absorption chiller regarding the effects of all thermal masses on the key parameters of an absorption chiller namely heat duty of all main components, coefficient of performance, and the exergetic efficiency was conducted by Iranmanesh and Mehrabian [13]. To validate the dynamic model, the results predicted from the dynamic simulation were compared with the steady-state results and relative errors for all cases were calculated. The results showed that thermal masses of main components have a minor effect on the heat transfer rate to/from low-pressure components (evaporator and absorber) while thermal masses of main components have a major effect on the heat transfer rate to/from high-pressure components (generator and condenser). It was also deduced that the thermal mass of the condenser has the highest effect on the heat duty of the generator and condenser, coefficient of performance, and the exergetic efficiency. A dynamic analysis of a single-effect absorption chiller by considering the effect of quality on concentration and ignoring the effects of the thermal masses of all components was accomplished by Iranmanesh and Mehrabian [14]. Moreover, a transient analysis of exergy was also achieved to determine the exergetic efficiency in terms of time. The simultaneous differential equations were solved using the fourth order Runge–Kutta method. Based on the deduced results, coefficient of performance and exergetic efficiency both decreased and finally approached the steady-state values. However, the results were in a good agreement with the ones derived from the steady-state analysis. Furthermore, as the quality approached zero, the effect of concentration could be neglected. A new computer model for optimization and simulation of an absorption refrigeration cycle was presented by Karno and Ajib [15]. The results obtained from the simulation were confirmed with an experimental investigation. It was deduced that the solution is well suited to run the system at low temperatures. The absorption process in LiBr incline plate absorber and the effect of film flow rate and plate angle on the heat and mass transfer phenomena was studied by Karami and Farhanieh [16]. It was concluded that absorption mass flux and heat and mass transfer coefficient increased as the angle of the plate increased. The replacement of an adsorption chiller by a newly developed absorption chiller was accomplished by Albers [17]. Furthermore, a control strategy with simultaneously controlled hot and cooling water temperature was implemented successfully for both chillers. It was concluded that by applying the control strategy to the absorption chiller, the seasonal energy efficiency ratio SEER was above 0.75, electric efficiency was 35% higher and water consumption was reduced by 70%. The performance optimization of solar driven small-cooled absorption–diffusion chiller working with light hydrocarbons was achieved by Sayadi et al. [18]. An absorption/diffusion machine was simulated using various binary mixtures of light hydrocarbons as working fluids in combination with helium. It was deduced from the technical and economic study that collector surface area was 6 m2 while annually total costs were 1.6 €/kW h. Thermodynamic and economical analysis of cooling systems including hot-water single-effect, hot-water double-effect, or direct-fired double-effect absorption chillers were accomplished based on the first and second laws of thermodynamics by Avanessian and Ameri [19]. It was deduced that it was not economical to use a single-effect absorption chiller. The payback period in the case of using the direct fired double-effect chiller was also calculated. It was concluded that the payback period in the case of using the direct-fired double-effect chiller is 4.97 and 3.54 years based on the prices in Iran and Toronto, Canada, respectively. Although thermodynamic modeling and economical and exergy analysis of various absorption chillers were achieved but nothing was done to minimize the energy costs of absorption chillers.
In this paper, optimization of a double-effect LiBr–H2O absorption refrigeration cycle is conducted using the genetic algorithm. Various parameters have contribution in optimization of the double-effect absorption chiller. Before optimization of the absorption chiller is achieved, it is better to study the behavior of each parameter separately. So, the sensitivity analysis of different parameters such as the volume of the storage tank, collector area, and the mass flow rates of water passing through the generator and collector are investigated. The storage tank is simulated based on the stratification model. In this model, the temperature of each node is evaluated on the basis of the energy balance equation applied on the storage tank. Two objective functions are defined for the genetic algorithm namely the auxiliary energy and net profit. The genetic algorithm is developed to minimize the auxiliary energy while the net profit is maximized. To absorb the highest solar radiation, the evacuated tube collector is mounted at the monthly optimum angle in Kerman.
Section snippets
Assumptions of thermodynamic modeling of the double-effect absorption chiller
The schematic diagram of the solar absorption system is illustrated in Fig. 1. To reduce the rate of energy consumption, an evacuated tube collector and a storage tank are used in the cycle. The evacuated tube collector absorbs the solar radiation during the day and the storage tank can save the absorbed solar energy and returns it to the system when the solar energy is not available. The following assumptions are made to accomplish thermodynamic modeling of the absorption chiller:
- •
The cooling
Results and discussion
Four parameters are considered as input parameters for genetic algorithm, as listed in Table 2. Two objective functions are defined in this case namely the amount of auxiliary energy and net profit. The goal of the optimization is to minimize the amount of auxiliary energy while net profit is maximized. The range of input parameters for genetic algorithm is shown in Table 2.
To pursue the economical analysis, it is necessary to know the average life time of the collector which is assumed 5 years
Conclusion
This paper deals with the optimization of the solar double-effect absorption chiller using genetic algorithm. Various parameters can minimize the rate of energy consumption. To pursue this approach, two objective functions were defined to optimize the absorption chiller. The results obtained in this research are listed below:
- (1)
Selection of each point on the Pareto chart depends on initial costs and investment return period.
- (2)
The efficiency of the evacuated tube collector becomes maximum when the
References (24)
Theoretical modeling of a solar air conditioning system for domestic applications
Energy Conversion and Management
(1993)- et al.
Modeling and simulation of solar absorption system performance in Beirut
Renewable Energy
(1997) - et al.
Performance of a second generation solar cooling unit
Solar Energy
(1998) - et al.
Modeling and simulation of an absorption solar cooling system for Cyprus
Solar Energy
(2002) - et al.
Simulation and optimization of a LiBr solar absorption cooling system with evacuated tube collectors
Renewable Energy
(2005) - et al.
Simulation of solar lithium bromide–water absorption cooling system with parabolic trough collector
Energy Conversion and Management
(2008) - et al.
Design and performance of solar powered absorption cooling systems in office buildings
Energy and Buildings
(2009) - et al.
Design of a solar absorption cooling system in a Greek hospital
Energy and Buildings
(2010) - et al.
An innovative solar-driven directly air-cooled LiBr–H2O absorption chiller prototype for residential use
Energy and Buildings
(2012) - et al.
Recognition of the importance of using artificial neural networks and genetic algorithms to optimize chiller operation
Energy and Buildings
(2012)
Dynamic simulation of a single-effect LiBr–H2O absorption refrigeration cycle considering the effects of thermal masses
Energy and Buildings
New absorption chiller and control strategy for the solar assisted cooling system at the German federal environment agency
International Journal of Refrigeration
Cited by (24)
Optimizing the thermal performance of solar energy devices using meta-heuristic algorithms: A critical review
2023, Renewable and Sustainable Energy ReviewsCitation Excerpt :They found that the dynamic optimization technique improves outcomes in all the cities examined, at both the design and retrofit phases. The influence of various factors on supplementary energy was studied using a sensitivity assessment on a double-effect absorption chiller by Iranmanesh and Mehrabian [85]. For the NSGA-II algorithm, two-goal functions were taken into account: supplementary energy and total revenue from the solar framework.
Cost-effective and environmentally friendly solar collector for thermally driven cooling systems
2022, Applied Thermal EngineeringOptimization of design and operation of solar assisted district cooling systems
2020, Energy Conversion and Management: XCitation Excerpt :The authors found out that emission levels had decreased significantly, and the model solved in short computational times for different case studies. Iranmanesh and Mehrabian [18] optimized the design of a SAC system by developing a multi-objective functions optimization model to minimize energy consumption and maximize the profit of the system. The optimization was carried out on MATLAB software, and the non-dominated sorting genetic algorithm was used to optimize the system.
Optimum number and arrangement of evacuated-tube solar collectors under various operating conditions
2019, Energy Conversion and ManagementPerformance analysis and optimization of new double effect lithium bromide–water absorption chiller with series and parallel flows
2019, International Journal of RefrigerationExperimental and theoretical performance evaluation of evacuated tube collectors under mediterranean climate conditions
2018, Thermal Science and Engineering Progress