Investigation of an Innovative Cascade Cycle Combining a Trilateral Cycle and an Organic Rankine Cycle (TLC-ORC) for Industry or Transport Application

Abstract

An innovative cascade cycle combining a trilateral cycle and an organic Rankine cycle (TLC-ORC) system is proposed in this paper. The proposed TLC-ORC system aims at obtaining better performance of temperature matching between working fluid and heat source, leading to better overall system performance than that of the conventional dual-loop ORC system. The proposed cascade cycle adopts TLC to replace the High-Temperature (HT) cycle of the conventional dual-loop ORC system. The comprehensive comparisons between the conventional dual-loop ORC and the proposed TLC-ORC system have been conducted using the first and second law analysis. Effects of evaporating temperature for HT and Low-Temperature (LT) cycle, as well as different HT and LT working fluids, are systematically investigated. The comparisons of exergy destruction and exergy efficiency of each component in the two systems have been studied. Results illustrate that the maximum net power output, thermal efficiency, and exergy efficiency of a conventional dual-loop ORC are 8.8 kW, 18.7%, and 50.0%, respectively, obtained by the system using cyclohexane as HT working fluid at T_HT,evap = 470 K and T_LT,evap = 343 K. While for the TLC-ORC, the corresponding values are 11.8 kW, 25.0%, and 65.6%, obtained by the system using toluene as a HT working fluid at T_HT,evap = 470 K and T_LT,evap = 343 K, which are 34.1%, 33.7%, and 31.2% higher than that of a conventional dual-loop ORC.

1. Introduction

Heat recovery technologies have attracted increasing attention around the world due to the increase in fuel prices, rigorous emission regulations, and concerns about environmental problems associated with the use of existing energy systems [1,2]. A number of waste heat recovery technologies have been proposed, such as organic Rankine cycles (ORCs) [3,4], Rankine cycles [5], absorption/adsorption cycles [6], Kalina cycles [7], thermoacoustic engines [8], and thermofluidic oscillators [9,10]. Among the stated cycles, ORC-based technologies are one of the most promising solutions because of their relatively high efficiency and simple configurations [11,12]. ORC technologies have been widely investigated and can be used to recover heat sources such as the waste heat from internal combustion engines [13,14], solar energy [15,16], geothermal, and other industrial applications [3,17,18]. However, the heat transfer process of a conventional ORC system in the evaporator suffers from poor performance of temperature matching between the heat source and working fluid due to the phase-change evaporating process, causing relatively high exergy destruction and poor thermodynamic performance [19,20]. Using mixture fluids can potentially solve the stated problem due to the non-isothermal process in the evaporator [21,22]. However, it is difficult to develop the mixture fluids, and their thermodynamic performance need to be further investigated [23].

The trilateral cycle (TLC) was previously proposed, which has a better thermal match from the heat source to the working fluid than conventional ORCs [7,24,25]. The characteristic of the TLC is that the working fluid directly flows into the expander after being heated in the evaporator without any evaporating process. Therefore, the temperature difference between the working fluid and heat source in the evaporator can be smaller than that of a conventional ORC, leading to an overall smaller exergy loss in the evaporator [26]. Johann Fischer compared the performance of a TLC and an ORC for waste heat recovery [27]. Results showed the exergy efficiency of the TLC can be 14–29% higher than that of a conventional ORC under the same operational conditions. M. Yari et al. studied TLC, ORC, and Kalina cycle using an exergoeconomic analysis method [7]. Results indicated TLC has better power output performance than that of ORC and Kalina cycle, but that the device cost is mainly determined by the isentropic efficiency of the selected expansion machine. A comparison of different pure working fluids illustrated that adopting n-butane can reduce the cost of the TLC system. Ramon Ferreiro Garcia et al. [28] pointed out that within a range of relatively low operating temperatures, the TLC can obtain an overall thermal efficiency at 44.9% compared to 33.9% of a Carnot cycle under the same high and low-temperature heat reservoir conditions [28]. Cipollone et al. [29] conducted a thermodynamic analysis of the TLC system considering the potential to use rotary positive displacement expanders as the expansion device to achieve two-phase expansion of the TLC. The reported feasibility study mainly focusing on the effects of using pure and mixture fluids on volumetric expansion machines under 10 kW power output [29]. The importance of considering the built-in volume expansion ratio for the TLC system was addressed, and results suggested that a relatively large built-in volume expansion ratio is required to produce a great specific work [29], which means that in order to achieve overall good performance, volumetric expansion machines are generally not recommend for use with TLC systems.

Previous studies indicated the overall performance of the heat recovery system using the TLC system can be higher than that of a conventional ORC system under the same operational conditions. However, previously-reported research on the TLC is mainly focusing on the performance studies of a single loop system. A single-loop TLC system will require a large evaporator heat exchanger to effectively and efficiently recover high-temperature heat sources such as from industrial processes and the transport sector. Moreover, a single-loop TLC cannot be used for multi-heat source recovery, which has less flexibility compared to a dual-loop cycle or cascaded systems. Extensive research efforts have been made on the dual-loop ORC system, because the system can effectively and efficiently convert dual heat sources into useful power. Moreover, dual-loop ORC technology has the advantage of being able to comprehensively recover a single heat source with a high-temperature profile. Wang et al. [30] first proposed the concept of a dual-loop ORC and investigated the performance of the system recovering engine exhaust and coolant energy from a gasoline engine. The results showed that the absolute effective thermal efficiency for the objective engine was improved by 3~6% under all operating conditions. Zhang et al. [31] designed a novel dual-loop ORC to recover the waste heat of exhaust gas, intake air, and coolant energy from a light-duty diesel engine. Results indicated that the effective power of the diesel engine can be improved by 14~16% and 38~43%, corresponding to peak effective thermal efficiency and low load (and high speed) conditions. Other representative works have been conducted by Shu and Tian et al. [22,32,33] and Song and Gu et al. [34,35]. Results showed that the system performed better at a high operating load, and R1234yf was found to be the best working fluid. Using water as the working fluid of the HT loop, the system obtained maximum net output power (39.67 kW) and exergy efficiency (42.98%) [32,33]. Song et al. [35] conducted a parametric study based on a dual-loop ORC system. Key parameters including condensation temperature of the high-temperature cycle and working fluids for high-temperature and low-temperature cycles were considered. The results based on a simulation showed that the maximum net power output could achieve as high as 111.2 kW by the proposed dual-loop ORC, increasing the engine power by 11.2% with cyclohexane and R236fa as high-temperature and low-temperature working fluid, respectively [35].

In summary, previous studies have compared the TLC with other power generation cycles, and have proven the TLC has great potential for heat recovery. Meanwhile, the superiority of using dual-loop ORC has been revealed, which can be used to effectively recover multi-heat sources, although large exergy loss still exists in the evaporator of the dual-loop ORC system. Therefore, in this study, the TLC has been introduced into a conventional dual-loop ORC system to form an innovative cascade system with the purpose of adding flexibility of the TLC for the potential application of multi-heat source recovery, and improving the overall energy efficiency of a conventional dual-loop ORC system. This study aims to study the thermodynamic performance of the proposed TLC-ORC system and conduct a comparative investigation of the conventional dual-loop ORC system under the same operational conditions. The key factors of the systems investigated in this work include the net power output, thermal efficiency, exergy efficiency, HT working fluids, LT working fluids, and the evaporating temperature of HT and LT cycles.

2. System Description

Figure 1 shows the layout of the proposed combined TLC-ORC waste heat recovery system, which consists of a High-Temperature (HT) loop and a Low-Temperature (LT) loop. The schematic diagram of the TLC-ORC is the same as the conventional dual-loop ORC system illustrated in Figure 1. The TLC is used to replace the HT loop, as shown in the dashed black line. The condensation heat from the TLC is used to preheat the working fluid in the LT loop. In order to understand and compare the working process of the TLC-ORC and a conventional dual-loop ORC system, the T-s diagrams of two systems are respectively presented in Figure 2 and Figure 3. The other technical challenge that should be noted is the expansion machine to be used in the TLC system, which would be forced to operate under two-phase conditions. Recently, works have confirmed the feasibility of using a screw-type expander [27,36] or a reciprocating engine [37] for the TLC system.

A comparison of the working principle between the proposed TLC-ORC and a conventional dual-loop ORC is illustrated in Figure 2 and Figure 3. The HT cycle of the TLC-ORC system maintains the working fluid as the liquid state before entering the expansion machine, which can potentially achieve a better thermal match of the heat source conditions. The detailed working process of the TLC is explained as follows, and is shown in Figure 2a. In state 1, working fluid is in a saturated liquid state at temperature T₁ and vapor pressure p₁. Then, the pressure is increased to p₂ by pump 1 at the state 2 in the homogeneous liquid. The liquid-phase working fluid enters the heat exchanger where it is heated to boiling point at pressure p₂, which is state 3. The temperature T₃ is the boiling temperature at pressure p₂. Starting from state 3, the working fluid directly enters the two-phase expander. In the two-phase expander, the working liquid expands into the wet vapor region, and gradually reaches the wet saturated vapor state (state 4), as shown in Figure 2a. The authors would like to point out the dryness of the working fluid at state 4 is determined by the calculation after the expansion process and condensation pressure controlled by the condenser. The working fluid at state 4 can be either saturated vapor, the wet state, or within the two-phase region. The wet saturated vapor is condensed until it reaches state 1. For the LT cycle in the TLC-ORC system, LT working fluid starts at state 5 with saturated liquid. Then, it is pressured to state 2 with pressure p₆ by pump 2. The LT working fluid absorbs residual heat from HT working fluid in condenser 1, and continues to absorb the waste heat of exhaust in evaporator 2. At the outlet of evaporator 2 (state 8), LT working fluid is saturated vapor; then, it flows into expander 2 to convert thermal energy to effective work by expanding the process. At state 9, LT working fluid is low pressure and it is condensed to saturated liquid at state 5 by cooling water. The working process of a traditional dual-loop ORC is similar to that of the proposed TLC-ORC. The main differences are the evaporating and condensing processes of the HT working fluid. The detailed working process of conventional dual-loop ORC can be found in reference [38].

3. Selection of Working Fluids

In order to investigate the characteristics of the TLC-ORC, the effects of different working fluids for HT cycle and LT cycle on system performance should be investigated. The selection of ORC working fluids should consider the following criteria [32,39,40,41].

(1)

Safety. Working fluid should be less flammable and toxic, or corrosive to pipes and other devices.

(2)

Environmental impact. Subjected to rigorous regulations, working fluid should have small potential issues when leaking to the environment.

(3)

Chemical stability. Working fluid should have a high decomposition temperature when recovering the waste heat of high-temperature heat.

(4)

Thermo-physical properties. Working fluid should have high vapor density, low viscosity, proper critical temperature, and freezing temperature.

According to the stated selection principles, five HT working fluids, i.e., cyclohexane, toluene, benzene, water, and octane, and five LT working fluids, i.e., R134a, R124, R245fa, R600, and R236fa, are selected. Detailed parameters of selected working fluids are listed in

Table 1. Properties of the selected high-temperature working fluid [42].

Working Fluid	Molecular Weight (g/mol)	Normal Boiling Point (K)	Critical Temperature (K)	Critical Pressure (MPa)
Cyclohexane	84.16	353.9	553.6	4.075
Toluene	92.14	383.8	591.8	4.126
Benzene	78.11	353.2	562.1	4.894
Water	18.02	373.1	647.1	22.064
Octane	216.37	398.8	569.3	2.497

and

Table 2. Properties of the selected low-temperature working fluid [43].

Working Fluid	Molecular Weight (g/mol)	Normal Boiling Point (K)	Critical Temperature (K)	Critical Pressure (MPa)
R134a	102.03	247.1	374.2	4.059
R124	136.48	261.2	395.4	3.624
R245fa	134.05	288.3	427.2	3.651
R600	58.12	272.7	425.1	3.796
R236fa	152.04	271.7	398.1	3.200

.

4. Methodologies

4.1. System Modelling

The thermodynamic model described in the following part can be used to calculate the performance for the dual-loop TLC-ORC system and conventional dual-loop ORC system. A series of imperative assumptions for use of the simulation method are listed below.

(1)

Each thermodynamic process in the system operates at a steady state condition.

(2)

Pressure drops along the pipes and heat dissipation in all pipes are ignored.

(3)

The kinetic energy and potential energy of working fluids are negligible.

(4)

Environmental temperature and pressure are assumed to be 298 K and 101 kPa respectively.

(5)

The isentropic efficiency for two turbines is set as 0.85, while the isentropic efficiency for the two pumps is assumed to be 0.65 [27].

(6)

The condensation temperature of HT and LT cycle is assumed to be 358 K and 308 K, respectively [32].

(7)

The minimum pinch point temperatures of the liquid-liquid heat exchanger and liquid-gas heat exchanger are assumed to be 10 K and 30 K, respectively [32].

Performance evaluation and parametric analysis in this paper are based on the energy and exergy equations according to the first and the second thermodynamic laws. The definition of exergy at each state point is calculated by Equation (1), where 0 represents the base state; it is set to be the ambient parameters in this work.

$$E_i=m\left[\left(h_i-h_0\right)-T_0\cdot \left(s_i-s_0\right)\right]$$

(1)

For HT cycle, each thermodynamic process is described by the following equations.

▪ The process from state 1 to 2

$$h_2=h_1+\left(h_{2s}-h_1\right)/{\eta }_{isen,p}$$

(2)

$$W_{p,1}=m_{wh}\cdot \left(h_2-h_1\right)$$

(3)

$$I_{p,1}=\left(E_1-E_2\right)+W_{p,1}$$

(4)

$${\eta }_{ex,p,1}=\left(E_2-E_1\right)/W_{p,1}$$

(5)

▪ The process from state 2 to 3

$$m_{wh}\cdot \left(h_3-h_2\right)=m_e\cdot \left(h_{e,in}-h_{e,m}\right)$$

(6)

$$I_{evap,1}=\left(E_{e,in}-E_{e,m}\right)-\left(E_3-E_2\right)$$

(7)

$${\eta }_{ex,evap,1}=\left(E_3-E_2\right)/\left(E_{e,in}-E_{e,in}\right)$$

(8)

▪ The process from state 3 to 4

$$h_4=h_3-\left(h_3-h_{4s}\right)\cdot {\eta }_{isen,expa}$$

(9)

$$W_{expa,1}=m_{wh}\cdot \left(h_3-h_4\right)$$

(10)

$$I_{expa,1}=\left(E_3-E_4\right)-W_{exp,1}$$

(11)

$${\eta }_{ex,expa,1}=W_{exp,1}/\left(E_3-E_4\right)$$

(12)

▪ The process from state 4 to 1

$$m_{wh}\cdot \left(h_4-h_1\right)=m_{wl}\cdot \left(h_7-h_6\right)$$

(13)

$$I_{cond,1}=\left(E_4-E_1\right)-\left(E_7-E_6\right)$$

(14)

$${\eta }_{ex,cond,1}=\left(E_4-E_1\right)/\left(E_7-E_6\right)$$

(15)

For LT cycle, mathematical equations for each thermodynamic process are described as follows.

▪ The process from state 5 to 6

$$h_6=h_5+\left(h_{6s}-h_5\right)/{\eta }_{isen,p}$$

(16)

$$W_{p,2}=m_{wl}\cdot \left(h_6-h_5\right)$$

(17)

$$I_{p,2}=\left(E_5-E_6\right)+W_{p,2}$$

(18)

$${\eta }_{ex,p,2}=\left(E_6-E_5\right)/W_{p,2}$$

(19)

▪ The process from state 7 to 8

$$m_{wl}\cdot \left(h_8-h_7\right)=m_e\cdot \left(h_{e,m}-h_{e,out}\right)$$

(20)

$$I_{evap,2}=\left(E_{e,m}-E_{e,out}\right)-\left(E_3-E_2\right)$$

(21)

$${\eta }_{ex,evap,2}=\left(E_8-E_7\right)/\left(E_{e,m}-E_{e,out}\right)$$

(22)

▪ The process from state 8 to 9

$$h_9=h_8-\left(h_8-h_{9s}\right)\cdot {\eta }_{isen,expa}$$

(23)

$$W_{expa,2}=m_{wl}\cdot \left(h_8-h_9\right)$$

(24)

$$I_{expa,2}=\left(E_8-E_9\right)-W_{expa,2}$$

(25)

$${\eta }_{ex,expa,2}=W_{expa,2}/\left(E_8-E_9\right)$$

(26)

▪ The process from state 9 to 5

$$m_c\cdot \left(h_{11}-h_{10}\right)=m_{wl}\cdot \left(h_9-h_5\right)$$

(27)

$$I_{cond,2}=\left(E_9-E_5\right)-\left(E_{11}-E_{10}\right)$$

(28)

$${\eta }_{ex,cond,2}=\left(E_{11}-E_{10}\right)/\left(E_9-E_5\right)$$

(29)

The definitions of the parameters for the system performance are described as the following equations.

$$W_{net}=\left(W_{expa,1}-W_{p,1}\right)+\left(W_{expa,2}-W_{p,2}\right)$$

(30)

$$Q_{in}=m_e\cdot \left(h_{e,in}-h_{e,out}\right)$$

(31)

$${\eta }_{th,total}=W_{net}/Q_{in}$$

(32)

$$E_{in}=\left(E_{e,in}-E_{e,in}\right)+W_{p,1}+W_{p,2}$$

(33)

$$E_{out}=\left(E_{11}-E_{10}\right)+W_{expa,1}+W_{expa,2}$$

(34)

$${\eta }_{ex,total}=W_{net}/\left(E_{in}-E_{out}\right)$$

(35)

The performance parameters, including the net power output, thermal efficiency, exergy efficiency, and component exergy destruction for the proposed TLC-ORC and conventional dual-loop ORC are compared, considering the effects of HT working fluids, LT working fluids, the evaporating temperature of HT cycle, and the evaporating temperature of LT cycle. The initial conditions are the same when calculating performances of both systems, and the main parameters can be found in

Table 3. Main parameters used in the calculation.

Parameters	Value and Unit
The inlet temperature of the cooling water	298 K
Condensing temperature of HT working fluid	358 K
Condensing temperature of LT working fluid	308 K
The inlet temperature of the exhaust	573 K
The outlet temperature of the exhaust	383 K
Mass flow rate of exhaust	804 kg/h
Evaporator pinch point temperature difference	30 K
Condenser pinch point temperature difference	10 K
Expander isentropic efficiency	0.85
Pump isentropic efficiency	0.65

. The isentropic efficiency of expanders for the proposed TLC-ORC and conventional dual-loop ORC is set to be the same as those of reference [27], in which a study was conducted to compare the performance of TLC and the basic ORC. The flow chart of the modelling and simulation process is shown in Figure 4.

4.2. Model Validation

To validate the thermodynamic model used in this paper, the performance of TLC and ORC are compared with the reference data under the same parameters.

Table 4. Comparison of TLC results between the present computation and reference [27].

Parameters	Q56 (kW)	ηth (%)	V3 (L/s)	CHC (kW/K)
Case III [27]	4581	21.83	6.24	19.85
Present model	4429	22.58	6.02	19.15
∆ (%)	3.32	3.44	3.53	3.52
Case IV [27]	5862	17.06	9.53	34.14
Present model	5614	17.81	9.10	32.69
∆ (%)	4.23	4.40	4.51	4.25

shows the results comparison of TLC between this paper and that reported by Fischer J. [27].

Table 5. Comparison of ORC results between the present computation and reference [44].

Parameters Unit	PORC (kW)	ηORC (%)	Pcond (kPa)	Pvap (K)	Tvap (K)	mf (kg/s)	∆h3-4 (kJ/kg)
Benzene	341.4	19.70	19.6	2000	494.6	2.690	130.5
Benzene [44]	349.3	19.86	19.6	2000	494.5	2.737	130.5
∆ (%)	2.26	0.81	0	0	0.02	1.72	0
R11	278.6	15.76	147.9	3835.9	461.6	7.873	41.1
R11 [44]	290.3	16.58	147.9	3835.9	461.0	7.487	41.9
∆ (%)	4.03	4.94	0	0	0.13	5.15	1.91

shows the results comparison of TLC between this paper and Vaja et al. [44]. Good agreement between the present study and results from the reference were observed. Therefore, the thermodynamic model used in the present study is reliable and can be used to conduct further calculations.

5. Results and Discussion

5.1. Effect of T_HT,evap Using Different HT Working Fluids

The effects of T_HT,evap on performances of two systems under different HT working fluids have been studied. T_LT,evap is fixed at 343 K and LT working fluid is R245fa. Figure 5 shows the comparative results of net power output between the two systems under different T_HT,evap. For TLC-ORC, net power output increases with the increase of T_HT,evap.

Table 6. Variation of heat consumed by the HT cycle of novel dual-loop TLC-ORC system.

THT,evap (K)	Cyclohexane (kW)	Toluene (kW)	Benzene (kW)	Water (kW)	Octane (kW)
440	45.79	45.84	45.79	45.87	45.86
455	45.75	45.82	45.74	45.85	45.84
470	45.71	45.80	45.69	45.83	45.82
485	45.65	45.76	45.64	45.81	45.80
500	45.59	45.72	45.56	45.78	45.77
515	45.51	45.68	45.48	45.74	45.73
530	45.42	45.62	45.39	45.69	45.68

shows the variation of heat consumed in the HT cycle of TLC-ORC system with different HT working fluids under various T_HT,evap conditions. A maximum net power output of 10.4 kW is obtained by a toluene-based TLC-ORC system with T_HT,evap = 530 K. It can be observed that the heat absorbed from the HT cycle by HT working fluids is almost the same under different T_HT,evap. For example, the heat consumed by the HT cycle of a TLC-ORC system using toluene as the HT working fluid only decreases by 0.8% when T_HT,evap increases from 440 K to 530 K, which means that the quantity of heat absorbed by LT cycle has little variation with T_HT,evap because total heat absorbed from exhaust and T_LT,evap are fixed parameters. The net power output of the LT cycle has little effect on the total net power output, and net power output of the HT cycle is the dominant factor with the change of T_HT,evap condition. Moreover, it can be observed that the net power output of TLC-ORC is not sensitive to the changes of HT cycle working fluids.

Regarding the net power output of dual-loop ORC shown in Figure 5, it has a completely different trend compared to that of TLC-ORC. For all the five HT working fluids, net power output first increases and reaches the peak point; then, it quickly falls off with the rise of T_HT,evap. The reason is related to the quantity of heat consumed by the HT cycle listed in

Table 7. Variation of heat consumed by the HT cycle of conventional dual-loop ORC system.

THT,evap (K)	Cyclohexane (kW)	Toluene (kW)	Benzene (kW)	Water (kW)	Octane (kW)
440	43.45	39.29	39.41	30.07	46.39
455	41.49	36.59	36.71	26.61	44.19
470	38.67	33.15	33.52	22.78	41.12
485	34.99	28.84	29.45	18.81	36.83
500	30.19	23.52	24.51	14.45	31.18
515	23.52	17.07	18.31	9.82	23.52
530	13.95	8.94	10.07	4.80	13.08

. Using toluene as an example, the quantity of heat absorption in evaporator 1 decreases by 77.2% when T_HT,evap increases from 440 K to 530 K. With the increase of T_HT,evap, specific enthalpy of HT working fluid increases while its mass flowrate declines at the inlet of the expander 1. Although heat absorbed by LT cycle increases, and the net power output of LT cycle increases slightly with the increase of heat input for LT cycle, it cannot compensate for the great reduction in the net power output of HT cycle. The maximum net power output of dual-loop ORC can achieve at 8.8 kW using cyclohexane as HT-cycle working fluid under T_HT,evap = 470 K, which is 14.8% smaller than that of TLC-ORC under the same operational conditions.

Variations in thermal efficiency of two systems using different HT working fluids with different T_HT,evap are shown in Figure 6. Results of the thermal efficiency of TLC-ORC show an upward trend as T_HT,evap increases for all HT working fluids. The difference of thermal efficiency among five HT working fluids is quite small, and the maximum value is obtained at 22.0% by TLC-ORC system using toluene as HT-cycle working fluid at T_HT,evap = 530 K. The thermal efficiency of the conventional dual-loop ORC shows the same trend as net power output. Because the total heat provided by the exhaust heat source is fixed, the trend of net power output with T_HT,evap determines the trend of thermal efficiency. Results indicate water as HT-cycle working fluid has the minimum thermal efficiency, while the maximum value has been obtained by cyclohexane as HT cycle working. The highest thermal efficiency is achieved at 18.7% by the conventional dual-loop ORC system using cyclohexane as HT-cycle working fluid with T_HT,evap = 470 K.

The exergy efficiency of the two systems has been studied, and the results are plotted in Figure 7. The exergy efficiencies of the TLC-ORC system are close to each other with the five HT working fluids. The maximum exergy efficiency among all the conditions is 57.9%, achieved by TLC-ORC system using toluene as the HT cycle working fluid. For the conventional dual-loop ORC system, under the same conditions, the system with cyclohexane as the HT-cycle working fluid achieves the best exergy performance, while that with water as HT-cycle working fluid shows the smallest exergy efficiency. The largest exergy efficiency is obtained at 50.0% by cyclohexane-based dual loop ORC system, which is 13.5% smaller than that of TLC-ORC system under the same operational conditions.

5.2. Effects of T_LT,evap Using Different HT Working Fluids

In order to evaluate the effects of T_LT,evap on the performance of two systems using different HT working fluids, the T_HT,evap was set at 530 K and the LT working fluid is selected to use R245fa. As shown in Figure 8, toluene has the largest net power output, while octane has the smallest at any T_LT,evap in the TLC-ORC system. It can also be seen that the net power output increases as T_LT,evap rises. The net power output of each HT working fluid differs slightly because the net power output of the HT cycle is dominant to the total net power output.

When the T_LT,evap = 368 K and toluene is used as the working fluid of the HT cycle, the net power output achieved from the system is 11.8 kW. In the conventional dual-loop ORC system, among the five HT working fluid candidates, the cyclohexane shows the best performance, while water performs worst under the same conditions. The maximum power output from the conventional dual-loop ORC system is 7.7 kW, achieved by the system using cyclohexane as HT-cycle working fluid, which is 34.6% smaller than that of TLC-ORC.

Figure 9 shows the change of thermal efficiency with T_LT,evap for different HT working fluids of two systems. As previously discussed, because total heat supplied from the exhaust is a constant value, the trend of thermal efficiency is identical to that of net power output for two systems. The thermal efficiency of the TLC-ORC, as shown in Figure 9, increases with the increase of T_LT,evap. The maximum thermal efficiency is obtained at 25.0% by the system using toluene as the HT-cycle working fluid at T_LT,eva = 368 K. The thermal efficiency of the conventional dual-loop ORC also increases with the increase of T_LT,eva as shown in Figure 9. Cyclohexane obtains a highest thermal efficiency, while water shows the lowest. The maximum value is achieved at 15.6% by the dual-loop system using cyclohexane as HT cycle working fluid, which is 37.6% less than the maximum thermal efficiency of the TLC-ORC.

Figure 10 shows the variation of exergy efficiency with T_LT,eva under different HT working fluids for two systems. In TLC-ORC system, the toluene-based system performs the best among the five work fluids, while benzene is the worst at any T_LT,evap. The power consumption of pumps for the benzene-based system is slightly larger than that of systems based on other HT working fluids; that is why the exergy efficiency of the benzene-based system ranked lowest. Overall, the maximum exergy efficiency is obtained at 65.6% by the toluene-based system with T_LT,evap = 368 K. In a conventional dual-loop ORC system, the cyclohexane-based system shows the maximum exergy efficiency is 43.4%, achieved at T_LT,evap = 373 K, which is 33.9% smaller than the maximum exergy efficiency of the TLC-ORC.

5.3. Effects of T_HT,evap Using Different LT Working Fluids

This section aims to compare the effects of LT working fluids and T_HT,evap on the output performance for two systems. Toluene is adopted as the HT working fluid of TLC-ORC, and cyclohexane is used for the conventional dual-loop ORC system due to its optimal performance, as pointed out in the previous sections. T_HT,evap is set at 343 K. Results of thermal efficiency and exergy efficiency are plotted in Figure 11 and Figure 12, respectively. The quantity of heat consumed by the LT cycle has little variation with T_HT,evap, and it accounts for a small part of the total heat absorbed from the exhaust heat source. Therefore, the output performance of the LT cycle has a small effect on the thermal efficiency of TLC-ORC system, which leads to a small difference of thermal efficiency among the five studied LT working fluids, as illustrated in Figure 11. R245fa has the best thermal energy, while R134a has the lowest under the studied conditions. Regarding the thermal efficiency of the dual-loop ORC, it first increases and reaches the peak point with the rise of T_HT,evap, then it falls off quickly. Although the mass flow rate of LT working fluids slightly increases with the rise of T_HT,evap, the increase of net power output in the LT cycle cannot compensate for the reduction of net power output in HT cycle, leading to thermal efficiency to first increase and then decline. The exergy efficiency of the two systems has been shown in Figure 12. The exergy of each system is almost the same under different operating conditions, except for the small difference of power consumption of the pumps. The R245fa-based system has the best exergy performance, while the R134a-based system has the lowest exergy efficiency.

5.4. Effects of T_LT,evap Using Different LT Working Fluids

To study the effects of LT working fluids and T_LT,evap on the output performance of the two systems, toluene is adopted as HT working fluid of TLC-ORC while cyclohexane is selected for dual-loop ORC. T_HT,evap is set as 530 K for both systems. The results of thermal efficiency and exergy efficiency under different LT working fluids and T_LT,evap. are shown in Figure 13 and Figure 14, respectively. As for TLC-ORC system, thermal efficiency and exergy efficiency increases with the increase of T_LT,evap. The highest thermal efficiency and exergy efficiency achieved by the system using R245fa as the LT-cycle working fluid range from 22.0% to 25.0% and from 57.9% to 65.6%, respectively. The lowest performance was obtained by the systems using R134a as the LT cycle working fluid, which ranges from 21.6% to 23.6% for thermal efficiency and from 56.0% to 59.8% for exergy efficiency. The maximum value is obtained at 25.0% by the TLC-ORC system using R245fa as LT-cycle working fluid at T_LT,evap = 368 K. The conventional dual-loop ORC system adopted R245fa as the LT-cycle working fluid, thereby achieving the highest thermal efficiency and exergy efficiency, respectively ranging from 12.4% to 16.4% and from 33.1% to 42.0%. The system using R134a as the LT-cycle working fluid shows the lowest value, ranging from 11.9% to 14.3% for thermal efficiency and from 31.3% to 36.4% for exergy efficiency. The maximum thermal efficiency of the conventional dual-loop ORC system is 16.4% using R245fa as LT-cycle working fluid at T_LT,evap = 373 K, which is 34.4% smaller than that of the TLC-ORC.

5.5. Analysis of Maximum Output and Component Performance

Effects of HT working fluids, LT working fluids, T_HT,evap and T_LT,evap on the TLC-ORC and the conventional dual-loop ORC are systematically studied and compared in previous sections. The maximum output performance for TLC-ORC is obtained by the toluene-based system at T_LT,evap = 368 K and T_HT,evap = 530 K. While for the dual-loop ORC, the best output performance is obtained by the toluene-based system at T_LT,evap = 343 K and T_HT,evap = 470 K. Detailed maximum output performance of the two systems can be found in

Table 8. Comparison of the maximum output between the dual-loop ORC and TLC-ORC.

Parameters	Dual-Loop ORC	TLC-ORC	Improvement (%)
Net power output (kW)	8.8	11.8	34.1
Thermal efficiency (%)	18.7	25.0	33.7
Exergy efficiency (%)	50.0	65.6	31.2

. The maximum net power output, thermal efficiency, and exergy efficiency of the conventional dual-loop ORC are 8.8 kW, 18.7% and 50.0%, respectively. In contrast, for TLC-ORC, the corresponding values are 11.8 kW, 25.0% and 65.6%, which are 33.5%, 33.5% and 31.1% larger than that of the conventional dual-loop ORC.

Figure 15 shows the exergy destruction and exergy efficiency of each component for both systems under the maximum output conditions, because the exergy destruction is the largest in the selected conditions. The top four exergy destruction among the system components in TLC-ORC are evaporator 1, expander 1, condenser 2, and expander 2, as illustrated in Figure 15a. And the exergy destruction of other components is quite limited, which means that further optimisation of TLC-ORC should concentrate on the previous four components. On the other hand, most of the components of the conventional dual-loop ORC system have large exergy destruction, except the two pumps. Compared with that of TLC-ORC, the exergy destruction in evaporator 1, evaporator 2, and condenser 1 are considerably larger than those of TLC-ORC, especially in condenser 1, leading to lower corresponding exergy efficiency shown in Figure 15b. Because TLC has better performance of temperature matching between HT working fluid and heat source condition, more exergy from exhaust heat source is transferred to the HT working fluid in evaporator 1 and smaller exergy destruction in HT cycle. However, the exergy destruction in expander 1 and expander 2 of the conventional dual-loop ORC is smaller than that of TLC-ORC. Because exergy in the HT cycle of TLC-ORC is large, and exergy efficiency of the two-phase expanding process in expander 1 is relatively smaller, accounting for the higher exergy destruction in expander 1. The exergy analysis conducted in this section has revealed the superior characteristics of the proposed TLC-ORC cycle, and pointed out the further optimization directions for the system.

5.6. Effects of the Isentropic Efficiency of the TLC Expander

In the previous sections, the isentropic efficiency of the expanders in the TLC-ORC and the dual-loop ORC system were all assumed to be 0.85, which aims to find the best output performance of the proposed TLC-ORC system, providing a performance comparison between the proposed TLC-ORC and conventional dual-loop ORC. However, the isentropic efficiency of the TLC expander operating in the two-phase region would be lower than that of a conventional ORC expander, which is used in the single-phase region. Therefore, it is worth investigating the effects of the isentropic efficiency of TLC expander on the system performance. The isentropic efficiency of the ORC expander is still set at 0.85, which has been widely adopted by other researchers to evaluate the thermodynamic performance of the ORC-based systems. Therefore, in this section, the maximum thermodynamic performance of the TLC-ORC under various TLC expander isentropic efficiencies was studied and compared to the dual loop ORC system, as illustrated in Figure 16. Results indicate that when the isentropic efficiency is 0.6, the maximum net power, thermal efficiency, and exergy efficiency of the proposed TLC-ORC are 8.3 kW, 17.7%, and 46.3% respectively, which are smaller than that of the conventional dual-loop ORC (8.8 kW, 18.7% and 50.0%). But when the isentropic efficiency increases to 0.65, the maximum net power, thermal efficiency, and exergy efficiency of the proposed TLC-ORC attain 9.0 kW, 19.2%, and 50.2%, respectively, which are slightly larger than that of the dual-loop ORC. In summary, when the isentropic efficiency of the two-phase expander is higher than 0.65, the TLC-ORC should be used; otherwise, the dual-loop ORC system should be considered.

6. Conclusions

In this paper, an innovative dual-loop ORC system introducing the TLC concept to effectively convert waste heat into useful power has been proposed. In order to compare the proposed TLC-ORC system performance with the conventional dual-loop ORC, the effects of HT working fluids, LT working fluids, T_HT,evap and T_LT,evap on the TLC-ORC and dual-loop ORC are systematically studied and compared, as well as exergy destruction and exergy efficiency of each component in two systems. The following main conclusions were obtained:

(1)

Compared to conventional dual-loop ORC system, net power output, thermal efficiency, and exergy efficiency of the proposed TLC-ORC system is not sensitive to the selection of HT working fluids. TLC-ORC using Toluene as the HT working fluid has the best energy and exergy performance, while cyclohexane-based the conventional dual loop ORC system using cyclohexane as the HT working fluid obtains the largest output.

(2)

Similar to that of conventional dual-loop ORC system, differences of output performance among five LT working fluids are quite small under various T_HT,evap, when the T_LT,evap is fixed. On the other hand, when the T_HT,evap is fixed, the LT working fluids affects the performance of both systems under different T_LT,evap. Results dictate that R245fa is recommended for use as the LT-cycle working fluid for both systems due to its highest energy and exergy efficiencies.

(3)

The maximum net power output, thermal efficiency, and exergy efficiency of the conventional dual-loop ORC are 8.8 kW, 18.7%, and 50.0%, respectively, obtained by the system using cyclohexane as HT working fluid at T_HT,evap = 470 K and T_LT,evap = 343 K. While for TLC-ORC, the corresponding values are 11.8 kW, 25.0%, and 65.6%, respectively, achieved by the system using toluene as HT working fluid at T_HT,evap = 530 K and T_LT,evap = 368 K, which are 33.7%, 34.1%, and 31. 2% higher than that of the conventional dual-loop ORC.

(4)

Under operating conditions with the largest output performance for two systems, for the TLC-ORC system, exergy destruction mainly occurs in expander 1, expander 2, evaporator 1, and condenser 2. In contrast, for the conventional dual-loop ORC system, exergy destruction in components is relatively large, except for that from the working fluid pumps. The exergy destruction of evaporator 1, evaporator 2, and condenser 1 in the TLC-ORC system are smaller than that of corresponding components in dual-loop ORC because TLC has better performance of temperature matching between HT working fluid and exhaust, leading to higher exergy efficiency in HT cycle.

(5)

The effects of the isentropic efficiency of TLC expander operating in the two-phase region have been studied. Results indicate that the proposed TLC-ORC has better overall performance than the dual-loop ORC when the TLC isentropic efficiency is higher than 0.65. Otherwise, the dual-loop ORC is recommended.