Analysis of combined cooling heating and power generation from organic Rankine cycle

and absorption system

Abstract
This paper focuses on the feasibilities of energy, economic and environment of a method
tenhance an ORC (organic Rankine cycle) efciency by CCHP (combined cooling heating
and power) generation from an absorption system for reducing the ORC condenser
temperature. A projection of a 25 kW R245fa ORC integrated with a 20 kW LiBr water
absorption unit was considered. The experimental data of both units were generated as
performance curves and used to nd out the suitable operating conditions. It could be found
that the ORC with the absorption system gave higher total efciency compared with the
normal ORC. The ORC efciency could be increased around 7%, with 15 C of cooled water
temperature supplied from the absorption system. But, in term of the economic result, a LEC
(levelized electricity cost) of the modied system was around 0.0891 USD/kWh, which was
higher than that of the normal system at around 0.0669 USD/kWh. In term of the
environmental impact, a released carbon dioxide intensity of the new unit was lower than the
normal unit at around 0.203 and 0.216 kg CO2 eq/kWh, respectively.

1. Introduction

ORC (Organic Rankine cycle) is a cycle that uses an organic working uid as a
working uid instead of water. Since the boiling point of the organic uid is lower than that
of water then it could be applied with various kinds of low temperature heat sources such as
geothermal energy, solar energy, biomass energy and waste heat. Today, rising of fossil fuel
prices and the environmental aspects on fossil fuel combustion, a signicant market for the
ORC is open and there is a challenge to develop an appropriate scale to meet both economic
and environmental needs.
Many studies on the ORC technique were reported. Walraven et al. [1] studied a
system optimization of ORCs cooled by air cooled condensers or wet cooling towers and
powered by low temperature geothermal heat sources. The results showed that it was
economic to use mechanical draft wet cooling towers instead of air cooled condensers. Suna
and Li [2] presented the ORC heat recovery power plant using R134a as working uid
mathematical models to evaluate and optimize the plant performance. Thawonngamyingsakul
and Kiatsiriroat [3] used a solar water heating system with a climate of Thailand to generate
and supply heat to ORC system. Thermo economic analysis was used to analyze the system
performance for CHP (combined heat and power) generation similar to other reports on
CCHP (combined cooling heat and power) [4-6].
Selection of ORC working uid was also an interesting topic and various literatures
[7-14] were reported. R-134a and R-245fa were always the recommended working uids.
Moreover, improvement of the ORC efciency by reducing condensing temperature such as a
design of high performance heat exchanger [15] and design of the condensation temperature
with respect to the expander characteristics [16,17] were carried out.
From the above mentioned literature review, it could be found that many studies
reported about the ORC applications. Techniques to enhance the ORC efciency such as
selection the suitable working uid and the optimal design were represented. It could be
noted that technique of reducing the working uid temperature at the ORC condenser by
using an absorption chiller did not represent in the recent literatures.
An interesting approach, a method to enhance the ORC efciency by CCHP
(combined cooling heating and power) generation from the absorption system for reducing
the ORC condenser temperature is considered. The experimental results of each technology
are performed to evaluate the optimal integrating system performance.

The aims of this study are as follows:

1. To evaluate performance curves of the ORC and absorption systems based on the testing
results.
2. To nd out systematic determination of optimum design parameters of the integrating unit.
3. To analyze economic result of the normal and new systems in term of a LEC (levelized
electricity cost).
4. To analyze environmental impact of the normal and new systems in term of a carbon
dioxide intensity.

2. System description

The operating principle of the ORC system is manifested in Fig. 1, hot uid as heat
source at temperature around 90 -120o C enters a boiler to heat and vaporize the working uid
at point 1 h and the outlet temperature of heat source is dropped at around 15-20 o C at point 2
h. The super heating working uid exiting the boiler at point 1 is expanded in a screw
expander to produce a mechanical work at point 2. An expander generator set is used in this
study. The expander operates at around 8000 rpm for feeding power and driving a reduction
gear box. The output of the gear box is around 3000 rpm and directly drives an induction
generator. The expander generator set is constructed to be a semihermetic screw type, where
the expander is integrated with the generator and installed inside a common casing to avoid
from vapor leak off. A lubrication system is used to reduce friction of a rotor (point 2), which
consists of oil and vapor separator (point 3), lter and oil pump (points 4-5). The presence of
lubrication system has lower the maintenance costs [18,19]. Lubricant and the vapor working
uid are separated at the oil and vapor separator (point 6). After that, the vapor is then
condensed in a condenser by cooling water at temperature around 25-35 o C to a low pressure
(P) as the subcooled working uid at (point 7). The uid at liquid state is compressed by a
refrigerant pump as a multi stage centrifugal pump to a high pressure (P High Low) at (point
8) and the new cycle restarts.
Fig. 2 shows the operating procedure of the combining unit, which the absorption
system is used to reduce the working uid temperature at the ORC condenser. Heat source
supplies heat to the ORC boiler (point 1 h) and sends to a generator of the absorption system
(point 2 h) at temperature around 90-120 o C and 70-90o C, respectively. Released heat
temperature at the generator will drop down to be around 60-75o C (point 3 h). A binary
liquid mixture consisting of a volatile component (absorbate) and a less volatile component
(absorbent) is obtained at the generator. The binary mixture (weak solution) is heated and part
of the absorbate boils at a high pressure (P High,Ab ) and temperature (T) at point 1a. The vapor
absorbate condenses in a condenser1 (TG) to be liquid at point 2a. After that, the absorbate in
liquid phase is throttled to a evaporator at point 3a of which a low pressure (P Low,Ab ) is lower
than that of the condenser1. The evaporator is heated by cooling water at temperature around
10-20o C (point 3c), after that cooling water will drop down temperature to be around 5-15 o C
(point 4c). In this study, cooling water comes from a cooling tower of the ORC system
(points 1c-2c), which is the cascade connection. This technique could be increased the
cooling capacity and decreased the cooling water temperature of the integrating system. The
absorbate at the evaporator is boiled to be vapor at point 4a and enters an absorber.
Meanwhile, the strong solution from the generator, at point 8a is sent through a heat
exchanger and a pressure reduction valve at points 9a and 10a, respectively, into the absorber
at the low pressure. In the absorber, the strong solution absorbs the absorbate vapor to be the
weak solution again. This liquid mixture leaves the absorber at point 5a at a medium
temperature (T) around 40-50o C, which is similarly the condenser temperature (T). The weak
solution at point 6a is compressed by a solution pump to the high pressure (point 6a) through
the heat exchanger (point 7a) into the generator again and new cycle restarts.
Fig. 2 also shows the concept of combined cooling heating and power, which waste
heat recovery from the ORC boiler is the useful heating mode. In cooling part, cooled water
from the evaporator of absorption chiller is combined with power generation to reduce the
ORC condenser and the ORC efciency could be enhanced.

3. Materials and methods

3.1. The ORC test rig

In the past, several working uids were considered with the ORC cycle such as
Hydrochlorouorocarbon (HCFC), Hydrouorocarbon (HFC) including of mixture
refrigerants etc [7-14]. In this study, a HFC-245fa refrigerant was selected because of low
pressure, high temperature, non corrosive, non ammable, low toxicity and friendly
environment. R-245fa was carried in a 25 kW e ORC prototype as shown in Fig. 3. The
descriptions of each ORC component were presented in Table 1.
In the testing procedure, water from a hot water generator was heated by a set of
diesel burner at heating capacity range of 200-600 kW. The required outlet water temperature
could be automatically adjusted with an uncertainty around 0.5 o C. The maximum water
temperature could be reached 120o C in the closed loop system. Hot water temperature
entering the ORC boiler were considered at 4 operating conditions (around 90, 95, 100 and
105o C), while the cooling water temperature entering the ORC condenser was xed at
around 28o C. The testing results of each operating condition were continually recorded every
1 min for 1 h at steady state condition. After that, the average value of the recorded data was
evaluated to carry out the ORC performance curve.
The ORC prototype was tested and measured by temperature, pressure and electrical
power sensors. Signal data form sensors were transferred to PLC programmed controller
(FATED programmed controller, FBs-CB55) and recorded into the PLC memory. T type
thermocouple at a precision range of 0.1o C [21], pressure transducer (WIKA, type A-10) at
a precision range of 0.25% [22] and current transformer (Shinohawa Electric, type YAL-1)
at a precision range of 1% [23] were used to measure temperature, pressure and electrical
power, respectively.

3.2. The bsorption test rig

 In the present study, a 20 kW water lithium bromide (LiBr) absorption chiller as
shown in Fig. 4 was used to nd out the thermal performance. In the testing procedure, a set
of electrical heater 30 kWe at accuracy of the controlled temperature 0.5o C was used to heat
water in a 1500 L storage tank. The descriptions of each absorption component were
presented in Table 2. For measurement, temperatures of the each working uid were
monitored by a T type thermocouple at a precision range of 0.1o C [21]. The output signal of
thermocouple was interfaced to a personal computer through a data logger and software of
TSUS Instruments at a precision range of 0.5% [24].
The electrical power consumptions of the each component were measured and
recorded by a power logger of KYORITSU [25] at an accuracy 0.2%. The recording time of
power logger was set matching with the temperature recorder.
In the next part, the constructed of the ORC and absorption units are tested its thermal
performance. The objective of this experiment is to nd out a performance curves of each
system, which is the correlation between the input parameters and the system efciency.
Projection of the optimal integrating unit by using the both performance curves is presented.
The system performance of the CCHP unit could be predicted under various operating
conditions. Moreover, this technique could be decreased complication of the fundamental
simulation.

4. Results and discussion

4.1. Thermal performance of the ORC system

The impact of hot uid temperature on the 25 kWe and the system performance were
investigated by setting the inlet hot uid temperature varying between 90 and 120 C, while
the cooling uid temperature was xed at around 28 C.

The testing result of the ORC system was shown in Table 3.It could be found that the cycle
efciency ( eff orc) of the R-245fa ORC system was lower than 5%, when hot water
temperature was lower than 90 C. While, the efciency could be increased to be over 8%,
when hot water temperature was higher than 97 C. This effect came from a vapor volume of
working uid entering the expander, which was a high value at the low temperature. But its
volume was nearly constant at the high temperature as shown in Fig. 5. Table 3 also shows
the supporting data of the expander inlet pressure (Phigh) and temperature (T1), which were
gradually constant at temperatures of 116, 107.8 and 97 C, respectively, and corresponding
with the cycle efciency. The average cycle efciency of the 3 operating points was 8.44%,
which corresponded with the above mentioned literature review [7-17] . Thus, in this study, 3
sets of testing data at temperature higher than 97 C were used to evaluate the performance
curve of the ORC system.

The correlation of the cycle efficiency (eff orc) and the different temperature between heat
source and heat sink (Thw,i Tcw,i) as shown in equation (1) and Fig. 6. This linear
correlation was relationships of the related data between the ORC efciency and input
variables, which were water temperatures of the water entering at the condenser and the
evaporator. It could be seen that the cycle efciency increased, when the temperature
difference increased, which followed the Carnot efciency concept. The cycle efciency
decreased, when the temperature difference of heat source and heat sink decreased. This
performance curve was used to predict the thermal performance, when the ORC system was
combined with the absorption system.

4.2. Thermal performance of the absorption system

The impact of heat source temperature on the 20 kW absorption unit and the system
performance were investigated by setting the inlet heat source temperature varying between
70 and 90 C.

Table 4 shows the average heating capacities of the absorption element. It could be seen that
the supplied heat at the generator was around 25 kW. The useful cooling capacity at the
evaporator was around 16 kW, while the rejected heats at the condenser and absorber were
around 22 and 21 kW, respectively.

Table 4 also shows the average temperatures of working uid, water entering and leaving the
main components of the absorption system. The cooled water leaving the evaporator was
around 15 C, which converts from the heat source temperature around 80 C at the generator.
The water LiBr concentrations at around 0.53 and 0.45%LiBr as the strong and weak
solutions based on %LiBr, respectively, were used to circulate between generator and
absorber. The experimental result also shown the average COP (coefcient of performance)
of the absorption prototype was around 0.66. This result was nearly the COP of the YAZAKI
absorption machine at around 0.7 [27].
When the experimental data at the heat source temperature varying between 70 and 90 C
wereanalyzed the absorption performance curve. The empirical correlation of the COP ab and
the different temperature term of (TG,i - TC1)/(TA,i - Te) was found in linear form as shown
in equation (2) and Fig. 7.

In this study, the performance curves of the ORC and absorption systems were used to
evaluate the optimal integrating unit by the projection method in the next part.

4.3. Projection of the optimal integrating system

In the projection process of the optimal integrating condition, the operating parameters and
the cycle efciency were predicted. The initial conditions of the projection method were
shown in Table 5.

Table 6 shows the projection results. It could be seen that the cycle efciency of the normal
ORC system (efforc,nirmal) was around 8.08% at the cooling water (T1C) around 32.22 C,
which was the typical water temperature of cooling tower under the climate of Thailand. The
normal ORC system obtained heat source (Qb) around 247.55 kWand converted heat to
power (WOorc,normal )of 20kW. For the modied system, after the same among of heat
source (Q)was used to generate electricity, low grade heat leaving the ORC boiler was
supplied to the absorption system to produce the cooled water temperature (T4c) around 15
C. The modied system could be decreased the cooling water (T1C) to around 24.87 C,
which the cooling tower was cascade connection with the absorption unit. In addition, the
electrical power (Worc,CCHP) of 21.39 kW was generated at the cycle efciency
(efforc,CCHP) around 8.64%.

From this result, it could be found that when the ORC condenser temperature decreased, the
ORC cycle efciency increased. Therefore the cycle efciency of the modied ORC system
could be increased around 6.97% compared with the normal ORC system. This result was
corresponded with the literature studies of (1,15-17), when the ORC condenser temperature
was reduced, the ORC efciency was enhanced. Moreover, when the evaporator of the
absorption unit was installed with the ORC condenser in cascaded connection, the cooling
capacity of the integrating unit could be increased around 70%. Thus the cooling water
temperature of the integrating unit could be decreased lower than the normal ORC unit,
which was the main advantage point of this technique.
4.4. Economic result

In economic assessment, the commercial absorption machine cost [26,28] as shown in Fig. 8
was used to evaluate the capital cost of the absorption system at capacity (Qe) around 166.50
kW (around 48 TR) as shown in Table 6. For the ORC cost, the present cost of the ORC
power plant varies between 2500e3000 USD/kWE [29-33] as given in Table 7. From the
table, it could be stated that the capital cost of a microscale ORC power plant at capacity
lower than 50 KW is around 2500 usd/ kw. Therefore this investment cost was used to
consider the economical assessment.

From the abovementioned literature review, payback period was selected to represent the
economic result of the ORC power plant. Because electricity rate of heat source was
specied. But in this study, heat source of the ORC system did not xe. Thus, the LEC
(levelized electricity cost) as presented in the studies of Thawonngamyingsakul and
Kiatsiriroat [3] and our former work [34] was selected to analyze in this study. Moreover, the
criteria for considering electricity price of renewable power in Thailand at payback period 10
y [35] was used to analyze the LEC. The initial conditions of economic evaluation were
shown in Table 8.

Table 9 shows comparison of economic results in term of the LEC factor, which based from
the operating condition of an only one geothermal power plant of Fang District, Chiang Mai
Province, Thailand. Thus the both LEC costs of this present work should be compared with
the electricity cost (Feed in Tariff, FiT) of renewable power plant in Thailand [37], which
unfortunate for the geothermal FiT does not promote. Because the only one geothermal ORC
power plant in Thailand was managed by the government (Electricity Generating Authority of
Thailand, EGAT).

From the simulation results, it could found that net electrical power production of the normal
and modied ORC system were around 16.68 and 17.07 kW, respectively. The ORC net
power (WORC,net) of the normal ORC unit came from the ORC cycle power (Worc)
subtracted by the electrical power consumptions of the cooling pump (Ebora model CMB/E 3
T, ow rate 100e280 L/min) and the fan motor of cooling tower (Model BKC 80 RT, fan
motor 1.5 hp) at 2.2 and 1.12 kW ORC, respectively. In case of the new unit, the electrical
power consumptions of the solution pump was considered in the absorption cycle at 78 We
(DAB in-line pump, model VSA 65/130, owrate 0.6 - 3.7 m3/h) [25]. When, the both ORC
units operated at 24 h/d and 350 d/y [36], the new unit could generate electricity higher than
the normal unit around 8600 kWh/y. For the LEC results, the CCHP unit was around 0.0891
USD/kWh, which was higher than the normal unit around 0.0222 USD/kWh.

When these costs were compared with the FiTs of renewable power plant in Thailand [37]. It
was found that the both LECs in Table 9 were lower than the FiTs of biomass and waste
(integrated waste management) at 0.096 USD/kWh, waste (land ll) at 0.171 USD/kWh and
biogas (waste water / waste material) at 0.115 USD/kWh. While the both LECs were
similarly with the FiT of biogas (energy plants) at 0.085 USD/kWh.

4.5. Environmental assessment

Energy, economic and environment impacts were considered in this study. For environmental
assessment, the electrical power consumption of the ORC process was investigated, which
consisted of refrigerant pump, oil pump, cooling pump and fan motor of cooling tower. In
addition, the power of solution pump was considered in the modied unit. Carbon dioxide
intensity of electricity of Thailand [38] at 0.6093 kg CO2 eq/kwh was used to estimate the
released CO2 from operating process of the ORC system. It could be found the normal unit
consumed the electrical power of 49,644 kWh/y and released the carbon dioxide intensity at
around 30,248 kg CO eq. While, the new unit could generate electricity higher than that of
the normal unit. Thus the carbon dioxide intensity from the output product of electricity 1
unit of the modied unit was lower than the normal unit of 0.203 and 0.216 kg CO2eq/kWh,
respectively.

5. Conclusions and recommendations

From the study result, it could be concluded that the method to enhance the ORC efciency
by combined cooling heating and power generation from the absorption system for reducing
the ORC condenser temperature was effective. The performance curves of each technology
were performed by using the experimental results for predicting the system performance.
Projection of the integrating system represented the better efciency, when the ORC
condenser temperature was decreased. The ORC efciency could be increased around 7%,
when the cooled water temperature at 15 C was supplied from the absorption system. But, in
the economic result, it could be found that electricity cost in term of LEC of the modied
system was higher than the normal system of 0.0891 and 0.0669 USD/kWh, respectively.
While, the environmental impact, the released carbon dioxide intensity of the new unit was
lower than the normal unit at around 0.203 and 0.216 kg CO eq/kWh, respectively.
For the future study, varying the ORC condenser temperature should be studied to nd out the
suitable cooled water temperature from the absorption system. The comparison effect of the
ammonia-water and water LiBr solutions of the absorption system should be investigated.
Moreover, different types of refrigeration system such as adsorption system and ejector
refrigerator should be studied.

Acknowledgments

The author would like to thank the School of Renewable Energy, Maejo University and
Center of Excellent for Renewable Energy, Chiang Mai University and the Ofce of the
Higher Education Commission, Thailand under the National Research University Program,
Chiang Mai University for supporting testing facilities.