DESIGN ΟPTIMIZATION OF RANKINE CYCLE SYSTEMS FOR WASTE HEAT RECOVERY FROM PASSENGER CAR ENGINES

Size: px

Start display at page:

Download "DESIGN ΟPTIMIZATION OF RANKINE CYCLE SYSTEMS FOR WASTE HEAT RECOVERY FROM PASSENGER CAR ENGINES"

Lorena Potter
6 years ago
Views:

DESIGN ΟPTIMIZATION OF RANKINE CYCLE SYSTEMS FOR WASTE HEAT RECOVERY FROM PASSENGER CAR ENGINES INTERNATIONAL FORUM AUTOMOTIVE THERMAL MANAGEMENT 2017 7-8

1 DESIGN ΟPTIMIZATION OF RANKINE CYCLE SYSTEMS FOR WASTE HEAT RECOVERY FROM PASSENGER CAR ENGINES INTERNATIONAL FORUM AUTOMOTIVE THERMAL MANAGEMENT February 2017, Michigan, USA Vincent LEMORT 1, Arnaud LEGROS 1, Olivier DUMONT 1, Mouad DINY 2 1 University of Liège, Belgium 2 PSA Peugeot Citroën, France

2 IntroducRon Context o TransportaRon represents a significant part of world energy consumpron and CO2 emissions. 250 o Worldwide regularons enforce to decrease the CO2 emissions of passenger cars o Increasing the efficiency of ICE is one of the mid- term strategies CO2 emission (g/km) Europe USA China Japan Mexico Brasil 2

IntroducRon Context Combus,on engine global efficiency can be increased by valorizing waste heat o Almost 2/3 of fuel energy is lost in exhaust heat and engine cooling circuit o Waste heat

3 IntroducRon Context Combus,on engine global efficiency can be increased by valorizing waste heat o Almost 2/3 of fuel energy is lost in exhaust heat and engine cooling circuit o Waste heat sources differ by the amount and quality of energy available. o There exists different techniques of waste heat recovery Ø To produce power Ø To produce cooling effect Ø To produce hearng effect 3

4 Content of the presenta,on 1. IntroducRon 2. Comparison of waste heat recovery techniques for passenger cars 3. Design of a scroll expander 4. Tests on a prototype of components 5. Conclusions and perspecrves 4

THERMOPHOTOVOLTAIC Efficiency Costs Maturity Packaging o Focus on power producson (other technologies available for cooling: ejector,

5 WASTE HEAT RECOVERY TECHNIQUES Comparison based on li<erature 5 Source: Legros et al., MECHANICAL ELECTRICAL TURBOCOMPOUND TURBOCOMPOUND RANKINE CYCLE THERMOELECTRIC THERMOACOUSTIC JOULE CYCLE STIRLING CYCLE THERMOPHOTOVOLTAIC Efficiency Costs Maturity Packaging o Focus on power producson (other technologies available for cooling: ejector, sorpron) o Turbocompounding and Rankine cycle are the most promising technologies o TEG: lower produced power and low maturity o Thermo- acousrcs/thermo PV: low maturity o Joule/SRrling cycle: large volume 5

6 WASTE HEAT RECOVERY TECHNIQUES Comparison based on simula,on Rankine cycle model o 3- zone evaporator model o Models calibrated based on experimental data o Condenser not modeled since condensing temperature is maintained constant by control 6

7 WASTE HEAT RECOVERY TECHNIQUES Comparison based on simula,on TEG and TC models o Thermal resistances network o TEG efficiency funcron of the figure of merit ZT.! = 1 +!" 1 1 +!" +!!"#$!!!"!!!"!!"#$!!!" o Regression model for the turbine efficiency and displaced mass flow rate o Model parameters idenrfied based on a turbocharger turbine 7

8 WASTE HEAT RECOVERY TECHNIQUES Comparison based on simula,on o WHR systems models are connected to a vehicle model o AddiRonal weight is taken into account o Back- pressure not considered o Power output is used to drive an electrical motor to boost the ICE 8

electrical power [W] Average electrical power [W] o Introducing a limit on the

9 WASTE HEAT RECOVERY TECHNIQUES Comparison based on simula,on o Turbocompound yields the best results if engine back- pressure not taken into account Average electrical power [W] Average electrical power [W] o Introducing a limit on the back- pressure sharply decreases the turbocompound power. Limit on the back- pressure [mbar] 9

10 WASTE HEAT RECOVERY TECHNIQUES Comparison based on simula,on Power [% of max value] 1 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0, Time [s] Turbocompound Rankine TEG Vehicle speed o Turbocompound yields the largest BSFC reducron if engine back- pressure not taken into account o RC shows less back pressure (depending on the evaporator hydraulic performance) => good compromise Speed [km/h] ReducSon of bsfc [%] o TC: largest power, less frequent Rme of use (mainly accelararon phases), highly sensirve to mass flow rate o RC more omen used (except during start up), mid power o TEG: lower power, almost awlways used, less sensirve to mass flow rate. NEDC Turbocharged engine WLTC5 Rankine Turbocompound Thermoelectric generator 10

11 Content of the presenta,on 1. IntroducRon 2. Comparison of waste heat recovery techniques for passenger cars 3. Design of a scroll expander for a Rankine cycle system 4. Tests on prototype of components 5. Conclusions and perspecrves 11

12 DESIGN OF A SCROLL EXPANDER Architecture of the Rankine cycle system 12

DESIGN OF A SCROLL EXPANDER Choice of expansion technology o Previous works stressed the advantage of scroll machines (design simplicity, reliability,

o However, no commercial high temperature scroll expander designed for such an applicaron is available on the market.

13 DESIGN OF A SCROLL EXPANDER Choice of expansion technology o Previous works stressed the advantage of scroll machines (design simplicity, reliability, promising performance, etc.) and water or ethanol (or a mixture) as working fluid. o However, no commercial high temperature scroll expander designed for such an applicaron is available on the market. o LubricaSng oil may be a major issue in ORC systems, especially in steam Rankine cycle (high operarng temperatures, oil separaron require bulky apparatus not comparble with mobile applicarons). An high- temp., oil- free scroll expander has been designed and prototyped. 13

14 DESIGN OF A SCROLL EXPANDER Which nominal point? o First step of the design is the sizing. o The World Harmonized Light Vehicles Test Cycle (WLTC) was applied to a 120 kw gasoline engine. o FrequenRal distriburon of power available in exhaust gases indicates that most of the Rme, the engine power in located in the first class (0-17). Frequential distribution [%] Exhaust gas available power [kw] o However, operarng condirons are highly transient. No nominal point can be easily defined and an oprmizaron of the scroll characterisrcs (mainly, the displacement) has to be conducted. 14

15 DESIGN OF A SCROLL EXPANDER Characteris,cs of the expander o Volume RaSo P W 1 W 2 V s r v,in.v s r v,in,ideal.v s V A built- in volume raro lower than the oprmal volume raro has been selected for compactness. That should also allow for a reducron of internal leakages. 15

DESIGN OF A SCROLL EXPANDER Characteris,cs of the expander o Displacement is oprmized based on quasi- starc simularon of the Rankine Cycle system (including a grey- box expander model) over the

16 DESIGN OF A SCROLL EXPANDER Characteris,cs of the expander o Displacement is oprmized based on quasi- starc simularon of the Rankine Cycle system (including a grey- box expander model) over the driving cycle. o The rotaronal speed is a funcron of the displacement and evaporarng pressure Low displacements or low pressures will yield high speed and the laqer is constrained Large displacements or high pressures will yield low speed and a larger impact of internal leakages 16

17 DESIGN OF A SCROLL EXPANDER Characteris,cs of the expander There exists an oprmal displacement maximizing the average power produced by the Rankine cycle system over the driving cycle. Average mechanical power [W] 17

o SucRon port cross- secronal area has been maximized by enlarging the clearance volume (which is

18 DESIGN OF A SCROLL EXPANDER Defining the geometry o A detailed scroll simularon model is used to define the exact geometry of the expander. o SucRon port cross- secronal area has been maximized by enlarging the clearance volume (which is not a drawback in expander mode) o Oil- free concept was selected: involute in coated aluminum and Rp seals in self- lubricarng material. Clearance volume 18

19 Content of the presenta,on 1. IntroducRon 2. Comparison of waste heat recovery techniques for passenger cars 3. Design of a scroll expander 4. Tests on prototypes of components 5. Conclusions and perspecrves 19

20 TESTS ON PROTOTYPES Descrip,on of the test rig o Open- loop steam Rankine cycle o Connected to a gasoline engine o Produced electricity dissipated in electric resistances 20

TESTS ON PROTOTYPES Evaporator o Two heat exchangers configurarons tested: counter- current and hybrid current Type of evaporator Counter Current (CC) Hybrid Current (HC) Mass [kg] x 1.77.

21 TESTS ON PROTOTYPES Evaporator o Two heat exchangers configurarons tested: counter- current and hybrid current Type of evaporator Counter Current (CC) Hybrid Current (HC) Mass [kg] x 1.77.x Volume [dm 3 ] y 0.83.y Water exchange area [m 2 ] z 0.36.z o o Performance expressed in terms of efficiency and pressure drops. Evaporator efficiency: raro between the actual and maximal heat transfer rates 21

22 TESTS ON PROTOTYPES Evaporator Large efficiencies Low pressure drop on the gas side Higher pressure drop on the water side (especially the CC) 22

o Maximum volumetric efficiency = 90% Parameter Value Swept volume 0.

23 TESTS ON PROTOTYPES Pump o o Gear pump Performance expressed in terms of isentropic and volumetric efficiencies o Maximum isentropic efficiency = 45% o Maximum volumetric efficiency = 90% Parameter Value Swept volume 0.5 cm 3 Max. pressure Max. mass flow (3000 RPM) Max. power consumpron 20 bar 20 g/s 105 W 23

24 TESTS ON PROTOTYPES Expander o Taylor made expander presented previously. o Two generarons of expanders have been built: V1 and V2 (20% reduced scroll lateral clearance). Parameter Value Swept volume 8 cm 3 Max. rotaronal speed RPM Max. temperature 250 C Pressure Volume raro Up to 20 bars ConfidenRal 24

25 TESTS ON PROTOTYPES Expander o o Performance expressed in terms of isentropic efficiency and filling factor. Max isentropic efficiency of 28% is achieved. o o Very low efficiency is explained by important internal leakages. For V2: oprmal rotaronal speed of 4000 rpm (antagonisrc effects of speed on mechanical losses and leakages) 25

TESTS ON PROTOTYPES Overall performance Evaporator 15000 W Expander 950 W

26 TESTS ON PROTOTYPES Overall performance Evaporator W Expander 950 W Efficiency= 25% Pump 65 W Efficiency= 35% Efficiency=5.9% Net power = 885 W > 600W 26

27 Conclusions and perspecrves ² A model- based comparison of different waste heat recovery techniques has been conducted, highlighrng the Rankine cycle system. ² A high temp. oil- free scroll expander has been designed, prototyped and tested. There is a large potenral of performance improvement. ² A pump and 2 evaporators have also been tested showing good performance. ² A reducron of BSFC of 5% on a NEDC is achievable. ² PerspecRves: o Tests with ethanol or with a mixture of water/ethanol will be conducted. o Test with oil will be conducted. o Other components will be tested. o Waste heat recovery on engine cooling loop will be invesrgated. 27

28 Thank you for your aqenron! 28

SIZING MODELS AND PERFORMANCE ANALYSIS OF WASTE HEAT RECOVERY ORGANIC RANKINE CYCLES FOR HEAVY DUTY TRUCKS (HDT)

SIZING MODELS AND PERFORMANCE ANALYSIS OF WASTE HEAT RECOVERY ORGANIC RANKINE CYCLES FOR HEAVY DUTY TRUCKS (HDT) Ludovic GUILLAUME 1 & co-workers : A. Legros 1, S. Quoilin 1, S. Declaye 1, V. Lemort 1,