Feasibility Study of ICE Bottoming ORC with Water/EG Mixture as Working Fluid

Size: px

Start display at page:

Download "Feasibility Study of ICE Bottoming ORC with Water/EG Mixture as Working Fluid"

Merry Payne
6 years ago
Views:

Available online at www.sciencedirect.com Energy Procedia 00 (2017) 000 000 www.elsevier.

Ziviani a,, Donghun Kim a, Swami Nathan Subramanian b, James E. Braun a, Eckhard A. Groll a a Ray W.

1 Available online at Energy Procedia 00 (2017) IV International Seminar on ORC Power Systems, ORC September 2017, Milano, Italy Feasibility Study of ICE Bottoming ORC with Water/EG Mixture as Working Fluid Davide Ziviani a,, Donghun Kim a, Swami Nathan Subramanian b, James E. Braun a, Eckhard A. Groll a a Ray W. Herrick Laboratories, Purdue University 177 S Russell Street, West Lafayette, IN, , USA b Eaton Corporate Research & Technology, Northwestern Hwy, Southfield, MI, 48076, USA Abstract To achieve the U.S. Department of Energys brake thermal efficiency (BTE) goal for Heavy Duty Diesel Engine (HDDE) technologies, Waste Heat Recovery (WHR) by means of Organic Rankine Cycle (ORC) systems has been selected as a suitable solution. The current relatively high return on investment period of such technology needs to be improved by significant cost reductions to realize benefits on WHR for mobile applications. The performance of the ORC system under dynamic loads relies on the choice of the working fluid, the efficiency of its components (mainly expander) as well as the control strategy that optimizes the operation. A novel ORC architecture is proposed that uses the engine coolant as the working fluid. In particular, a fraction of the engine coolant, which is a mixture of water and ethylene glycol, is employed as working fluid through the ORC to recover waste heat from EGR (Exhaust Gas Recirculation) and part of the tail pipe exhaust gases. At the inlet of the expander, the mixture has mixed-phase conditions and a fixed volume ratio expander is employed to generate power output that can be fed directly to the engine crankshaft. Heat rejection is accomplished through the spare capacity of the engine radiator, which avoids the need for a separate condenser. To evaluate the feasibility of such system architecture, a thermodynamic steady-state cycle model has been developed to predict the potential increase of BTE under different engine loads as well as to understand the ORC performance. Parametric studies are carried out by varying the system pressure ratio, the internal volume ratio of the expander and the mixture quality at the expander inlet. c 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the scientific committee of the IV International Seminar on ORC Power Systems. Keywords: ORC; Waste Heat Recovery; Heady-Duty Diesel Engine; Water-Ethylene Glycol mixture; 1. Introduction It is well known that Heavy Duty Diesel Engines (HDDEs) reject a considerable amout of energy to the ambient. In order to meet the U.S. Department of Energy (DOE) break thermal efficiency (BTE) goals [1], waste heat recovery Corresponding author. Tel.: ; address: dziviani@purdue.edu c 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the scientific committee of the IV International Seminar on ORC Power Systems.

2 2 D. Ziviani et al. / Energy Procedia 00 (2017) (WHR) by means of an organic Rankine cycle (ORC) has been identified by U.S. engine manufacturers as a viable solution. During the recent years, research on ORC systems applied to passenger and commercial vehicles has seen a rapid growth as such power cycle combines maturity and cost-effectiveness. For example, Amicabile et al. [2] carried out a design optimization of an ORC integrated into a heavy-duty diesel engine by considering both subcritical and supercritical cycle architectures. Ethanol and pentane were identified as suitable working fluid both in terms of power output as well as costs. The ORC solutions proposed were costing averagely $15,000 for a class 8 line-haul truck [2]. However, only the Exhaust Gas Recirculation (ERG) cooler was considered as heat source. A more comprehensive study to exploit waste heat in both exhaust gases and the engine coolant has been done by Chen et al. [3]. A novel confluent cascade expansion (CCE) ORC system has been proposed to improve more conventional dual-loop ORCs. The new architecture running with cyclopentane allowed to generate up to 8% more net power compared to the conventional dual-loop ORC. The break specific fuel consumption (BSFC) was reduced from 185 g/(kwh) to g/(kwh). Cost, complexity, environmental and safety issues are the major issues of ORC systems installed in vehicles. The return on investment period for the end customer is not highly attractive by using the current technology (3 to 4 years payback period). The successful commercialization of ORC systems is seeing a major impediment from OEMs. In this paper, an affordable ORC system is analyzed in order to obtain real benefits of WHR on the road and reduce the costs by 50% with a targeted pay-back period of 1.5 to 2 years. Nomenclature h specific enthalpy, J/kg ṁ mass flow rate, kg/s p pressure, Pa Q heat rate, W T temperature, C Ẇ power, W η efficiency, - T PP pinch point temperature difference, C ɛ effectiveness, - 2. ARC system description The novel ORC architecture proposed within the ARC project is based on using the engine coolant as the working fluid. The engine coolant is typically a water - Ethylene Glycol (EG) mixture with a mass fraction composition of [ ]. As shown in Fig. 1, a small portion (usually <0.5% of total mass flow rate) of the engine coolant in liquid-phase is pressurized by means of a pump (state 1 to state 2) and used to recover waste heat from the exhaust gas recirculation (EGR) system (state 2 to state 3) and exhaust tail pipe (state 3 to state 4). While absorbing heat, the water/eg mixture becomes a wet binary mixture as it undergoes through partial evaporation. The high pressure two-phase water/eg mixture is then expanded through a fixed-volume ratio expander (state 4 to state 5), which is able to handle two-phase conditions. Heat rejection is accomplished through the engine radiator, avoiding the need for a separate condenser for the ARC system. However, limitations arise concerning the maximum heat rejection rate. To ensure normal operation of the truck engine, the following constraints are taken into account: return temperature of engine coolant into the engine after EGR boiler; maximum engine coolant temperature at expander inlet; exhaust tail pipe boiler exit temperature. Such constraints are dictated by safety reasons, emission control and thermal stability of engine coolant.

3 D. Ziviani et al. / Energy Procedia 00 (2017) EGR HEX 3 2 Air in Water/EG Pump 1 Filter Fuel in Exhaust EGR loop Engine 5 4 Tail Pipe HEX Expander Exhaust to ambient Water/EG Loop Radiator Fig. 1. ARC system architecture for WHR from EGR and tail pipe exhaust gases. Temperature [ C] EG 0.00 EG 0.50 EG Specific Entropy [kj/(kg K)] Fig. 2. T-s thermodynamic plots for different concentrations of EG. A thermodynamic cycle model is developed to analyze such cycle architecture and to demonstrate the feasibility of using the engine coolant as working fluid to reach similar fuel economy benefits as conventional ORCs Water-EG mixture thermophysical properties The working fluid is a binary mixture of water and ethylene glycol. Few studies have been found about the estimation of thermodynamic and transport properties of such mixture [4,5]. As the mixture phase-change is an important aspect to both recovery heat and generate power output, the vapor-liquid equilibrium (VLE) conditions need to be obtained. VLE diagrams (or temperature-concentration diagrams) are used to demonstrate the concentration shifts within the liquid and vapor phases, as described in Section 2.2. REFPROP [6] is used to retrieve the thermophysical properties of the water/eg mixture. Figure 2 shows T-s diagrams obtained for pure water, pure EG and a water-eg mixture with a EG mass fraction of 0.5 which has been selected for further analyses. To be noted is that the properties close to the critical point are not defined. However, in temperature and pressure ranges of interest for the cycle calculations, no converge issues have been experienced.

4 4 D. Ziviani et al. / Energy Procedia 00 (2017) Table 1. Assumptions and constraints of the thermodynamic cycle model. Parameter Value Description Mixture concentration (mass fraction) [ ] Engine coolant concentrations T water EG,max, C Issues with thermal stability above 200 C p max, kpa 2000 Expander limitations Tail pipe HEX T PP, C 5 Design choice p cond, kpa variable Related to radiator operating conditions Tail pipe HEX T PP, C 5 Design choice Minimum expander inlet quality, Design choice η is,exp, Typical range for expanders [9] η is,pump, Design choice 2.2. Thermodynamic cycle model Based on the system architecture shown in Fig. 1, a steady-state cycle model has been developed to investigate the performance of the ARC. The heat inputs are determined from the engine operation. Furthermore, constraints on the maximum temperature of the coolant as well as recirculated exhaust gas temperature entering into the engine and tail pipe exhaust exit temperature are imposed to ensure safe operation of the engine as well as emission controls. The modeling assumptions and design constraints are listed in Tab. 1. The total heat rate available at the EGR can be quantified as: Q EGR,in = ɛ EGR ṁ EGR ( hegr,in h EGR,out ) where h EGR,in and h EGR,out are the inlet and outlet enthalpies of the EGR and they are fixed by the engine operating conditions. A heat exchanger effectiveness is applied to obtain the heat recovered by the coolant. The heat rate available from the exhaust tail pipe, Q TP,in, is defined analogously to Eq. 1. The effectiveness of the heat exchangers has been assumed to be The heat rejected by the radiator is calculated by Q cond = ṁ water EG h radiator (2) Note that if the heat rejection limitations are applied, the exit temperature of the radiator is imposed. Both the pump and the positive displacement expander have been modeled by assuming a constant value of the isentropic efficiency. For the expander, the internal volume ratio is accounted for to estimate the specific work during the expansion process [7]. The cycle performance and the benefits of the ARC system are quantified by defining an ORC thermal efficiency and Break Power (BP) improvement as: η ORC,net = ẆORC,net Q tot,in = Ẇexp Ẇ pump Q EGR,in + Q TP,in (3) (1) BP = ẆORC,net Ẇ engine (4) The model has been implemented in EES (Engineering Equation Solver) [8] coupled with the REFPROP library. Parametric studies are conducted by varying the independent variables, i.e. pressure ratio, expander isentropic efficiency and expander inlet temperature. Furthermore, the quality of the water-eg mixture at the expander inlet is also a degree of freedom that depends on the high side pressure and mass flow rate for the given heat sources. The amount of EG that evaporates directly influences the work that can be extracted from the expander. It is important to point out that the mixture water-eg presents thermal stability issues at temperatures above 200 C. As development studies are ongoing to improve the working temperature range, parametric studies are carried out in the present work up to 300 C to understand the potential of adopting engine coolant as an ORC working fluid.

5 Table 2. Nominal engine operating points considered for the analysis. D. Ziviani et al. / Energy Procedia 00 (2017) Parameter # 1 # 2 # 3 # 4 # 5 # 6 # 7 # 8 T EGR,in, C T TP,in, C Table 3. ARC model results for each engine operating condition. The results are obtained by fixing the condensing pressure at 150 kpa and a pressure ratio across the expander of 8. Parameter # 1 # 2 # 3 # 4 # 5 # 6 # 7 # 8 Expander inlet temperature, C Expander isentropic efficiency Net output power, kw ORC system efficiency, Total heat input, kw BP improvement, % Working fluid mass flow rate, g/s Expander specific volume ratio, Expander inlet mixture quality, Expander outlet mixture quality, Expander inlet EG concentration (vapor phase), Expander outlet EG concentration (vapor phase), Fig. 3. T-s thermodynamic plots of ARC system under engine operation # 2. The plot has been obtained with mixture concentration of [ ] and evaporating and condensing pressures of 1600 kpa and 200 kpa, respectively. 3. Results and Discussion In order to evaluate the performance of the novel cycle architecture, several engine operating conditions have been identified. The engine parameters are not presented in this paper due to confidentiality. However, the inlet temperatures of EGR and tail pipe heat exchangers are reported in Tab. 2. The cycle model has been used to simulate all the operating points. As previously mentioned, a number of constraints have been taken into account while running the simulations. During the first case study, the maximum temperature of the water-eg mixture has been set equal to 220 C. At higher temperatures, the mixture tends to decompose and potentially compromise the engine performance. A pressure ratio of 8 is imposed across the expander due to the maximum pressure that the considered expander technology can safely support. The pump isentropic efficiency is set equal to 0.6, whereas the expander isentropic efficiency is in the range from 0.6 to 0.8. This range is representative of the majority of positive displacement expander performance [9]. This condition is named scenario 1. An example of a thermodynamic cycle plot is shown in Fig. 3. Engine condition #2 has been used to generate the plot. To be noted is that for a mixture composition of [ ], the resulting cycle is a partial evaporating Rankine

6 6 D. Ziviani et al. / Energy Procedia 00 (2017) BP [-] η exp, is = 0.6 η exp, is = 0.8 #1 #2 #3 #4 #5 #6 #7 #8 (a) ORC Efficiency [-] η exp, is = 0.6 η exp, is = 0.8 #1 #2 #3 #4 #5 #6 #7 #8 (b) Fig. 4. Parametric study of ARC system with maximum expander inlet temperature of 220 C with scenario 1: (a) BP improvements; (b) cycle thermodynamic efficiency. BP [-] η exp, is = 0.6 η exp, is = 0.8 ORC Efficiency [-] η exp, is = 0.6 η exp, is = #1 #2 #3 #4 #1 #2 #3 #4 (a) (b) Fig. 5. Parametric study of ARC system with scenario 2 (reduced or without tail pipe heat recovery): (a) BP improvements; (b) cycle thermodynamic efficiency. Only four engine operating conditions are shown. cycle. The quality at the expander inlet represents a degree of freedom to be optimized, although the maximum temperature allowed for the mixture is fixed. At first, simulations are carried out for each engine operating conditions by maintaining the expander efficiency constant at 0.6. The main results are provided in Tab. 3. The cycle model is the exercised to evaluate the influence of the expander isentropic efficiency. The results of the parametric study are reported in Fig. 4. In particular, Fig. 4(a) shows the break power improvements for each operating conditions at two different expander isentropic efficiency values. To be noted is that when an isentropic efficiency of 0.6 is considered, the BP improvements are below 5% with the exception of engine operating conditions #6 and #7. Due to cycle constraints, the thermodynamic efficiency, reported in Fig. 4(b), presents limited variability. In this first analysis, no limitations have been imposed to the heat rejection rate at the condenser, i.e. an additional heat exchanger can be installed to handle the heat load at the condenser side. By introducing the heat rejection limits at the engine radiator, i.e. scenario 2, the BP improvement drops significantly, as shown in Fig. 5(a). However, the analysis seems to suggest that the cycle efficiency is less sensitive to the heat rejection limitations. This is partially due to the constraints imposed to the simulation, e.g., pressure ratio and maximum temperature. At this point, two constraints were relaxed to understand the potential of the ARC architecture. The maximum allowed temperature of water-eg mixture was raised up to 300 C and the maximum pressure ratio was set equal to

7 D. Ziviani et al. / Energy Procedia 00 (2017) BP [-] T exp, in = 220 C, r p = 8 T exp, in = 300 C, r p = 8 T exp, in = 300 C, r p = 10 #1 #2 #3 #4 #5 #6 #7 #8 (a) ORC Efficiency [-] T exp, in = 220 C, r p = 8 T exp, in = 300 C, r p = 8 T exp, in = 300 C, r p = 10 #1 #2 #3 #4 #5 #6 #7 #8 (b) Fig. 6. ARC system analysis with relaxed constrains on temperature and pressure values: (a) BP improvements; (b) cycle thermodynamic efficiency. Expander Specific Work [kj/kg] r p = Water Mass Fraction [-] Fig. 7. Effect of varying the water-eg mixture concentration on expander specific work output at fixed expander inlet temperature of 220 C. 10. The calculation have been performed once again for all the engine operating conditions by keeping the expander isentropic efficiency fixed at 0.8. The results are shown in Fig. 6(a) and Fig. 6(b). The upper curve of Fig. 4(a) corresponding to an expander inlet temperature of 220 C, pressure ratio 8 and expander isentropic efficiency of 0.8 is taken as reference case. By increasing both the maximum temperature and pressure, the BP increased appreciably allowing engine conditions from #3 to #7 to reach or exceed the 5% target. A maximum BP improvement of 7.8% was achieved for engine condition #7 with a cycle thermodynamic efficiency of It is interesting to observe the behavior of the mixture quality at the expander inlet. At 220 C and 1200 kpa, the mixture quality is approximately 0.44, while at 300 C and the same pressure, the mixture is superheated. The increased temperature upper limit leads to improve the expander specific work by up to 80%. In the current analysis the concentration of the water-eg mixture has been kept constant. However, it is interesting to evaluate the effect of increasing the water content. By maintaining an expander temperature limit of 220 C, the expander specific work output increases if the mass fraction of water increases, due to the fact that the quality of the mixture increases as well, as reported in Fig Conclusions In this work, a novel organic Rankine cycle for waste heat recovery within heavy-duty trucks that employs a water - Ethylene Glycol mixture has been proposed. A thermodynamic cycle model has been developed to investigate the potential improvements on the engine brake thermal efficiency. Simulation results showed that engine coolant can

8 8 D. Ziviani et al. / Energy Procedia 00 (2017) potentially used as working fluid but its employment is heavily conditioned by engine operating conditions, high temperature limitations and expander performance. The maximum BP improvement obtained was 6.94% for engine operating point #7. Although the parametric studies showed some potential for the ARC architecture, additional work is needed to allow the system to work at higher temperatures and pressures to compete with traditional ORC configurations. Furthermore, as positive displacement expanders are considered in this work, the feasibility of using water-eg mixture in the liquid phase as lubricant should be evaluated. Acknowledgements This material is based upon work supported by the Department of Energy Vehicle Technologies Program under Award Number DE-EE The DOE/industry funded project is entitled Affordable Rankine Cycle (ARC) Waste Heat Recovery for Heavy Duty Trucks. The Authors would like to acknowledge Brandon Rouse from PAC- CAR for his experimental work on the engine baseline data. The authors greatly appreciated the support of Dr. Eric Lemmon and Dr. Ian H. Bell from NIST addressing calculation of the thermophysical properties of water/eg mixtures. Finally, the authors would like to acknowledge Dr. Abhinav Krishna for his leadership and contributions during the first year of the project. Disclaimer This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. References [1] (Accessed on March 6, 2017).???? [2] Amicabile, S., Lee, J.I., Kum, D.. A comprehensive design methodology of organic Rankine cycles for the waste heat recovery of automotive heavy-duty diesel engines. Applied Thermal Engineering 2015;87: [3] Chen, T., Zhuge, W., Zhang, Y., Zhang, L.. A novel cascade organic Rankine cycle (ORC) system for waste heat recovery of truck diesel engines. Energy Conversion and Management 2017;138: [4] T., S., Teja, A.S.. Density, viscosity, and thermal conductivity of aqueous ethylene, diethylene, and triethylene glycol mixtures between 290 K and 450 K. J Chem eng Data 2003;48: [5] Dai, J., Wang, L., Sun, Y., wang, L., Sun, H.. Prediction of thermodynamic, transport and vapor-liquid equilibriuim properties of binary mixtures of ethylene glycol and water. Fluid Phase Equilibria 2011;301: [6] Lemmon, E.W., Bell, I.H., Huber, M.L., McLinden, M.O.. NIST Standard Reference Database 23: Reference Fluid Thermodynamic and Transport Properties-REFPROP, Version 9.1.1, National Institute of Standards and Technology [7] Macchi, E., Astolfi, M., editors. Organic Rankine Cycle (ORC) Power Systems. Woodhead Publishing; Chapter 12 - Positive displacement expanders for Organic Rankine Cycle systems, Lemort, V., Legros, A., pp [8] Klein, S.. Engineering Equation Solver, F-Chart Software [9] Imran, M., Usman, M., B-S, P., Lee, D.H.. Volumetric expander for low grade heat and waste heat recovery applications. Renewable and Sustainable Energy Reviews 2016;57:

Performance Benefits for Organic Rankine Cycles with Flooded Expansion

Purdue University Purdue e-pubs Publications of the Ray W. Herrick Laboratories School of Mechanical Engineering 6-2-2010 Performance Benefits for Organic Rankine Cycles with Flooded Expansion Brandon