Thermodynamic Optimization of a Combined Cooling and Power System Utilizing Industrial Waste Heat

Low-grade thermal energy (below 230 °C) accounts for about 60% of industrial waste heat and has significant recycling potential [1]. However, single utilization methods such as power generation, refrigeration cannot achieve deep utilization of this low-grade heat energy [2]. Therefore, studying the cascade utilization of industrial waste heat, particularly low-grade waste heat, is highly significant.

Efficient approaches to harness low-grade industrial waste heat include enhancing heat upgrading, generating electricity, and refrigeration [3]. However, these techniques are often limited by the temperature range of the waste heat, making complete utilization a challenge. Dividing the heat source into distinct temperature regions and utilizing them in a cascading manner is a common approach to address suboptimal utilization of waste heat and incomplete energy conversion [4]. For example, Mohammadi et al. [5] established a combined cooling, heating, and power system comprising a gas turbine, an organic Rankine cycle (ORC), and an absorption refrigeration cycle (ARC), achieving an exergy efficiency of 68%. Zhou et al. [6] introduced a cascade system combined with an ARC and ORC, investigating how critical parameters impact the system's thermodynamic performance. However, most of these studies focused only on the effect of key parameters on the system, neglecting the optimization of the system as a whole.

Additionally, since electricity is the highest-grade energy source, ORC power generation is highly valued in the field of industrial waste heat recovery. For ORC systems, most of the studies have focused on selecting the optimal working fluid. Emadi et al. [7] utilized genetic algorithms and neural networks to optimize 20 working fluids for a solid oxide fuel cell waste heat recovery system, revealing that R601 emerged as the optimal choice. Wang et al. [8] compared ORC and proton exchange membrane fuel cells using different working fluids, finding that a zeotropic mixture of R245fa/R123 (0.6/0.4) demonstrated the best thermodynamic performance. Wang et al. [9] employed a multi-objective genetic algorithm to optimize the zeotropic working fluids for a dual-loop ORC system using flue gas waste heat. The results indicated that cyclohexane/cyclopentane (0.2/0.8) and butane/pentane (0.65/0.35) performed best in the high-temperature and low-temperature loops, respectively. In these studies, zeotropic working fluids are considered an effective option for improving the thermodynamic performance of ORC systems. However, most existing studies focus on optimizing the mass fraction of a zeotropic working fluid with a predetermined composition. Few studies have simultaneously optimized both the composition and the mass fraction of the zeotropic working fluid.

In this study, a combined cooling and power (CCP) system is proposed for the deep utilization of low-grade industrial waste heat, including an ORC with zeotropic working fluids and an ABR. A particle swarm optimization (PSO) algorithm is employed to optimize its operating parameters as well as the composition and mass fraction of the ORC working fluid. Based on this, the thermodynamic performance of the optimized system is analyzed.

The proposed system is illustrated in Fig. 1. It comprises an ORC subsystem for power generation and an ABR subsystem for refrigeration. In this system, the heat source (flue gas) is divided into two sections: Sec. 1, the high-temperature section, serves as the heat source for the ORC subsystem, while Sec. 2, the low-temperature section, is used as the heat source for the ABR subsystem.

In this study, the working fluid of the ORC subsystem is considered as a binary zeotropic mixture. The specific components and mass fractions are obtained from the optimization (specific optimization method is given in Sec. 3), and the thermophysical properties of each alternative component (from Refprop 9.1) are listed in Table 1. Ammonia–water solution and lithium bromide–water solution are widely employed as working fluids in ABR. Compared to lithium bromide–water solution, ABR with an ammonia–water solution as the working fluid achieves lower refrigeration temperatures. Therefore, ammonia–water solution is selected as the working fluid for the ABR subsystem.

The whole flowchart of the PSO is shown as Fig. 2.

The base design parameters of the proposed system are listed in Table 3. The variation curves of the PSO fitness values under these parameter settings are shown in Fig. 3, and the optimal design parameters are listed in Table 4.

Based on the optimization results, the most suitable working fluid for the ORC subsystem is the R141b/R124 with a mass fraction of 0.625:0.375. Under this configuration, the ORC subsystem exhibits a specific net power output of 6.09 kW/kg_FG (net power output per unit mass flowrate of flue gas), accompanied by a thermal efficiency of 14.07%. Additionally, when operating at a refrigeration temperature of −5.9 ℃, the ARC subsystem demonstrates a specific refrigeration capacity of 27.65 kW/kg_FG and a COP of 0.4453. Consequently, the overall system showcases an exceptional equivalent specific output power amounting to 11.45 kW/kg_FG.

In addition, to demonstrate the advantages of zeotropic working fluids for the ORC subsystem of the proposed system, the ORC subsystem with zeotropic working fluid was compared with the ORC subsystem using a single-component working fluid, as shown in Fig. 4. It is worth noting that the ORC subsystem with a single-component working fluid employs the same optimization methodology (PSO) to ensure a fair comparison. Using zeotropic working fluids in the ORC results in an increase in net power output by 0.43 kW/kg_FG and thermal efficiency by 0.99% compared to using R141b alone. Furthermore, when compared to an ORC employing R124 as its working fluid, the use of zeotropic working fluid leads to a significant improvement in net power output by 1.53 kW/kg_FG and thermal efficiency by 3.53%. These findings demonstrate that zeotropic mixtures substantially enhance the thermodynamic performance of the ORC subsystem.

By focusing on the exergy flow within the system, the exergy analysis provides valuable insights into the distribution of “energy quality” losses in the thermodynamic system. The input exergy of the system originates from industrial flue gas and certain components, resulting in a total input exergy of 25.553 kW. The output exergy primarily comprises power generated by ORC and cooling capacity provided by ARC, totaling 11.230 kW. Consequently, the exergy efficiency stands at 43.9%. A comprehensive examination of exergy flow and loss within this system is presented below to provide a deeper understanding.

Figure 5 illustrates the exergy flow of the proposed CCP system. As depicted in the figure, the system's total input exergy amounts to 25.553 kW, with flue gas waste heat contributing 22.192 kW (86.8% of the total). The remaining exergy is supplied by pumps, distillers, and the non-freezing fluid in the ARC evaporator. After passing through the heat exchanger, the industrial flue gas delivers 9.535 kW of exergy to the power generation system and 8.816 kW to the refrigeration system. However, as this exergy traverses various heat exchangers, pumps, throttling valves, and other equipment, losses occur amounting to 14.313 kW (56.1% of the total input). The ORC and ARC exhibit exergy efficiencies of 53.2% and 35.7%, respectively.

Figure 6 illustrates the proportion of exergy loss in the primary equipment of the system. The ORC contributes 39.15% of the overall exergy loss, while the ARC accounts for 60.85%. Notably, heat exchanger equipment plays a dominant role in causing exergy losses. Specifically, within the ORC, the condenser and evaporator contribute to 73.7% of the total exergy loss. Similarly, heat exchanger equipment in the CCP system is responsible for 56.0% of the total exergy loss. Therefore, further enhancements for the proposed system should primarily focus on optimizing and designing more efficient heat exchangers.

This study introduces an innovative CCP system that effectively harnesses waste heat from industrial processes, specifically utilizing flue gas waste heat through a cascade utilization approach. A PSO algorithm is employed to optimize the proposed system, specifically focusing on the composition and mass fraction of the zeotropic working fluid in the ORC subsystem. The main findings are as follows:

1. The optimized ORC subsystem of the proposed system achieves a specific net power output of 6.09 kW/kg_FG and a thermal efficiency of 14.07%. Meanwhile, the ABR subsystem attains a specific refrigeration capacity of 27.65 kW/kg_FG and a COP of 0.4453.

2. The zeotropic working fluid significantly enhances the thermodynamic performance of the ORC subsystem. Compared to ORC systems using pure R141b and pure R124, the ORC subsystem of the proposed system experiences a net power output increase of 0.43 kW/kgFG and 1.53 kW/kgFG, respectively.

3. In the ORC subsystem, heat exchanger exergy losses account for 73.7% of the total exergy losses. In the ABR subsystem, heat exchanger exergy losses account for 53.6% of the total exergy losses. Therefore, further enhancements for the proposed system should primarily focus on optimizing and designing more efficient heat exchangers.

This work was supported by the State Key Laboratory of Low-Carbon Smart Coal-Fired Power Generation and Ultra-clean Emission and National Natural Science Foundation of China (Grant No. 52022020).

There are no conflicts of interest.

The datasets generated and supporting the findings of this article are obtainable from the corresponding author upon reasonable request.