Simulation study on performance and emissions of a Small Direct Injection Diesel Engine fueled by Dimethyl Ether

Transcription

1 Simulation study on performane and emissions of a Small Diret Injetion Engine fueled by Dimethyl Ether Nguyen Lan Huong, *, Kanit Wattanavihien, Luong ong Nho and Pham Huu Tuyen 3 Mehanial Engineering Faulty, Vietnam Maritime University, 484 Lah Tray Street, Haiphong ity, Vietnam Internal ombustion Engine Researh Laboratory, hulalongkorn University, Phayathai Road, Patumwan, Bangkok, 33 3 Shool of Transportation Engineering, Hanoi University of Siene and TehnologyNo Dai o Viet Street, Hanoi, Vietnam *orresponding Author: nlhuongkdt@gmailom Abstrat Reently, Dimethyl ether () has been onsidered as a potential alternative fuel for diesel engine This paper presents a simulation study on performane and ombustion harateristi of a small diret injetion diesel engine fueled by The engine is modeled by AVL Boost and the model is verified by experiment Due to the differene in fuel properties, the engine power with is lower than that with diesel at full load, but emissions are improved On average with, the engine power redues by 555% To maintain the engine power, the fuel mass per yle must be inreased by 395% In this ase, smoke, O, and NO x of engine respetively redue 7597%, 7% and 7559% ompared with those of diesel engine Keywords: engine simulation, alternative fuel, dimethyl ether, exhaust gas emissions Introdution Nowadays, researh and utilization of renewable fuels in order to ensure energy seurity and redue pollution emissions are of interest in many ountries Among these fuels, Dimethyl Ether () is a friendly - environment fuel, easy to liquefy and suitable for use in diesel engines, hemial formula is H 3 -O-H 3, is a olorless organi ompound is in gaseous form at ambient pressure and temperature To inrease the energy density, is usually stored in liquid form under ompressed pressure of 7 to bar an be produed from a variety of raw materials suh as biomass, oal and natural gas and it is onsidered as a lean alternative fuel in near future Using for diesel engine may redue not only dependene on fossil fuel but also environmental pollution is not a nature produt but a syntheti produt whih is produed through either dehydration of methanol or a diret synthesis from syngas is now attrating great attention as

2 an energy soure for the st entury beause of its available soures, exellent physial and hemial properties and exellent storage properties Some experimental investigations were onduted on diesel engine to larify how injetion harateristis affet the engine performane and exhaust emissions [, ] Most of the results showed that emissions redued signifiantly when fueled by The aim of this paper is to performane and emissions of fuelled diesel engine A Kubota diesel engine is modeled with diesel fuel and by using AVL Boost software Model desription ombustion model The ombustion in diesel engine an be onsidered by two proesses: premixed ombustion and mixing ontrolled ombustion proesses[8] dq dq dq () total M PM Q total : Total heat release over the ombustion proess [kj] Q PM : total fuel heat input for the premixed ombustion [kj] Q M : umulative heat release for the mixture ontrolled ombustion [kj] - Ignition delay model: The ignition delay is alulated using the Andree and Pahernegg [8] model by solving the following differential equation: di T T id UB ref () Q ref As soon as the ignition delay integral I id reahes a value of (=at α id ) at the ignition delay id is alulated from id id SOI I id : ignition delay integral [-] T ref : referene temperature = [K] T UB : unburned zone temperature [K] Q ref : referene ativation energy, f(droplet, diameter, oxygen ontent, ) [K] id : ignition delay α SOI : start of injetion timing [ dega] α id : ignition delay timing [ dega] Premixed ombustion model: A Vibe funtion is used to desribe the atual heat release due to the premixed ombustion [8]: dq PM Q a m y e id y dq ( m) PM m a y (, ) (, ) M omb F M f m Q f k V (3) Q PM : total fuel heat input for the premixed ombustion = m fuel,id PM m fuel,id : total amount of fuel injeted during the ignition delay phase PM : premixed ombustion parameter α : premixed ombustion duration= id- PM-Dur PM-Dur : premixed ombustion duration fator m : shape parameter m= a : Vibe parameter a= 69 - Mixing ontrolled ombustion proess: In this regime the heat release is a funtion of the fuel quantity available (f ) and the turbulent kineti energy density (f ): (4)

3 with QM f( mf, Q) mf LV f ( k, V ) k m F, I turb E Diff w kin Oxygen, available Rate 3 V (5) omb : ombustion onstant [kj/kg/deg A] Rate : mixing rate onstant [s] k : loal density of turbulent kineti energy [m /s ] m F :vaporized fuel mass (atual) [kg] LV: lower heating value [kj/kg] V: ylinder volume [m 3 ] α : rank angle [deg A] w Oxygen,available : mass fration of available Oxygen (aspirated and in EGR) at SOI [-] EGR EGR influent onstant [-] k m E kin : kineti jet energy [J] stoih EGR (6) turb : turbulent energy prodution onstant [-] m F,I : injetion fuel mass (atual) [kg] Diff : Air Exess Ratio for diffusion burning [-] m stoih :stoihiometri mass of fresh harge [kg/kg] Emission model NOx Formation Model 6 reations introdued in Table, whih are based on the well-known Zeldovih mehanism are taken into aount: Table NO x formation reations Stoihiometry Rate TA i a T ki k, i T e R N +O= r =k N O NO+N R O +N= r =k O N NO+O R 3 N+OH= NO+H r 3 =k 3 OH N R 4 N O+O=NO+ NO r 4 =k 4 NO O R 5 O +N =N O+ O r 5 =k 5 O N R 6 OH+N = N O+H r 6 =k 6 OH N All reations rates r i have units [mole/m 3 s] the onentrations i are molar onentrations under equilibrium onditions with units [mole/m 3 ] The onentration of N O is alulated aording to: T N O N O T e P The final rate of NO prodution/ destrution in [mole/m 3 s] is alulated as[7]: r NO Post ProMult KinetiMult with AK r r AK AK 4 r r r 3 NO, at 4 NO, equ Post Pr omult 4, AK4 r5 r6 r (7) (8) O Formation Model O formations of two following reations given in Table are taken into

4 aount: Table O formation reations Stoihiometry Rate R O+OH= O +H r R 676 e O OH O +O= r O+O O O T e 55 T The final rate of O prodution/ destrution in [mole/m 3 s] is alulated as [7]: r r r (9) o onst, with O at O, equ 3 Soot formation model Soot formation is desribed by two steps inluding formation and oxidation The net rate of hange in soot mass m soot is the differene between the rates of soot formed m sootform and oxidized m sootox [7] with dm dm dm dmsoot dmsoot soot form ox soot form n dm fuel p yl Aform e p diff ref Ta form Tave Ta ox () () soot ox n O Tave Aox msoot e har p Oref () - A form : soot formation fator [-] - A ox : soot oxidation fator [-] - τ har : harateristi mixing time [ A] - m fuel : mass of fuel burned [kg] - T a-form : ativation temp: soot formation [K] - T a-ox : ativation temp: soot oxidation [K] p n3 - T ave : average in-ylinder temperature [K] - p yl /p ref : normalized in-ylinder press [-] - p O /p Oref : normalized oxygen partial press [-] - n, n, n 3 : model fator [-] 3 Fuel desription In this study, Dimethyl ether is defined as gas that inludes O, O, H, H Properties of and an be speified as follow: Molar mass: 46 kg/ mol 7kg/mol Lower heating value : 76 MJ/kg 45MJ/kg Stoihiometri A/F ratio: arbon/: total mass ratio : 5 86 Oxygen/ total mass ratio: Heat transfer model The heat transfer to the walls of the ombustion hamber, ie the ylinder head, the piston, and the ylinder liner, is alulated from equation [9] Qwi Ai i ( T Twi ) (3) Where Q wi - wall heat flow, A i - surfae area, α i - heat transfer oeffiient, T - gas temperature in the ylinder, T wi - wall temperature Heat transfer oeffiient (α i ) is usually alulated by WOSHNI Model, The Woshni model published in 978 for the high pressure yle is summarized as follows [9]:,,8,53 w 3 D p T v T,8 [ D m ( p p, )] p, V, (4) Where = u / m, =,34 for DI engines, D - ylinder bore, m - mean piston speed, u - irumferential veloity, u = πdnd/6, V D - displaement per ylinder, p,o - ylinder pressure of the motored engine (bar), T,

5 BSF (g/ kwh) - temperature in the ylinder at intake valve losing (IV), p, - pressure in the ylinder at IV (bar) 5 Modelling diesel engine Kubota RT Kubota RT Engine is a single horizontal ylinder, four-stroke, naturally aspirated, water ooled, DI The engine speifiation is shown in Table 3, and the model is built by AVL Boost software (Fig ) Table 3 Speifiations of the engine Rating output Maximum torque Bore / Stroke Swept volume ompression ratio Injetion pressure Nozzle number x orifie diameter/ mm Stati injetion timing kw at rpm 4 Nm at rpm 97/96 mm 79 m 3 8: bar 4 x 3 5 A BTD engine was run at full load ondition and engine speed varied from rpm to rpm [3] Fig Experimental lay out 4 Results and disussion 4 Model validation The model has been validated by experiment in ase of using onventional diesel and It showed that the torque as well as fuel onsumption between simulation and experiment in ase of diesel mathed quite well: on average the differene in torque and fuel onsumption was about 7 % and 9 %, respetively Thus, it is possible to use this model to simulate the engine with fuel(fig3, Fig4) Ex Si Fig engine Kubota RT model 3 Experiment for model validation In order to validate the model, the experiment was arried out on engine test bed (Fig) The diesel engine was ouple to a dynamometer whih an test the engine The Fig 3 Torque with diesel 7 Speed(rpm) Fig 4 Fuel onsumption with diesel Ex Si

6 BSF (g/kwh) BSF (g/kwh) In ase of, the torque as well as fuel onsumption between simulation and experiment are similar, on average the differene in torque and fuel onsumption was about 7 % and 3 % (Fig5, Fig6) Ex Si 7 Fig 5 Torque with 7 Fig 6 Fuel onsumption with 4 Simulation results of engine performane and emissions when fueled by 4 Engine performane With the same fuel mass per yle of 444 (g/yle), torque of engine when fueled by is smaller than that fueled by diesel (Fig 7)This is due to the lower heat value of as ompared to diesel (Q H_ = 886 kj/kg, Q H_diesel = 48 kj/kg) Ex Si 3 7 Fig 7 Engine performane when using diesel and (444g/y) To maintain the engine power, it is neessary to inrease the fuel mass per yle by 395% (Fig 8) and as a result, the fuel onsumption orrespondingly inreases (Fig 9) Fig 8 Engine power with (444g/yle) and diesel(67 g/yle) 3 7 Fig 9 Fuel onsumption with diesel (444g/yle) and (67g/yle) 4 Effets of injetion timing on engine performane The torque redues when inreasing injetion timing from to 9 rank angle (A) degree before top dead enter (TD) at rpm, 7rpm and rpm (Fig ) It is due to earlier injetion timing whih leads to inrease the peak of ylinder pressure and shift it to the left

7 NOx (ppm) O( ppm) Smoke (g/kwh) Pyl(Pa) Rate of Heat Release(J/deg) (Fig ) Injetion timing (deg) rpm 7 rpm rpm Fig Engine power versus injetion timing at different speed deg 5 deg 9 deg A (deg) Fig ylinder pressure at rpm When inreasing speed, the torque reahes maximum value with earlier injetion timing, max torque appears with injetion timing of 9 A degrees before TD Similar to diesel fuel, in ase of the higher engine speed the shorter evaporation and mixing time Therefore, the fuel needs to be injeted earlier in order to have suffiient time and more omplete ombustion During the ombustion proess, the peak of rate of heat release in ase of ours at 38 A degree with the value of 46J/degree that is about 53% higher than that of diesel (Fig ) This may be due to the better vaporization of that leads to better mixture with air and faster ombustion 3 Rate of Heat Release A (deg) Fig Rate of heat release at rpm 43 Exhaust emission The results show that the exhaust emission hanges when using Fig 3 shows the signifiant redution in smoke when using, up to 848 % at rpm Averagely, smoke redues by 7597% over speed range The variation of O emission is depited in Fig 4 The O emission with is lower than that with diesel On average, the O emission redues by 7% The redution of NOx emission is shown in Fig 5 with average value is 7559% Fig 3 Smoke emission O emission 7 NOx emission Fig 4 O emission 7 speed (rpm) Fig 5 NOx emission

8 4 onlusions Performane and emissions of fueled Kubota diesel engine are simulated by AVL Boost software The model is verified by experiment The results show that: With the same fuel mass per yle, the engine power with is lower as ompared to that with diesel To maintain engine power, it is neessary to inrease the fuel mass per yle Similar to diesel fuel, the fuel needs to be injeted earlier when inrease speed in order to have suffiient time and more omplete ombustion ompared to diesel, redues quite learly smoke, NOx and O Referenes [] Yoshio Sato, Akira Noda and LiJun, Effets of Fuel Injetion harateristis on Heat Release and Emissions in a DI Engine Operated on SAE , [] Dimethyl ether as alternative fuel for I engine and vehile, Zhen HUANG, Xinqi QIAO, Wugao ZHANG, Junhua WU, Junjun ZHANG [3] Kanit Wattanavihien Implementation of in a Small Diret Injetion Engine st AUN- SeeedNet Regional Workshop on New and Renewable Energy, -3 Marh 9, at ITB, Bandung, Indonesia [4] Nguyen Lan Huong, Luong ong Nho, Pham Huu Tuyen Dimethyl Ether ()- An alternative fuel for diesel engine Journal of Transportation Deember- [5] Nguyen Lan Huong, Luong ong Nho, Pham Huu Tuyen Investigating Dimethyl ether () fuel systems for engine Journal of Transportation Marh- 3 [6] Nguyen Lan Huong, Luong ong Nho, Pham Huu Tuyen Simulation study on diesel engine fueled by Dimethyl Ether () sientifi onferene mehanial pneumati 3 [7] G D Errio, et al () Modeling the Pollutant Emissions from a SI Engine, SAE paper No --6 [8] Users guide- AVL Boost version Theory- AVL Boost version [9] GWoshni (967) A Universally Appliable Equation for the Instantaneous Heat Transfer oeffiient in Internal ombustion Engines SAE paper No 6793 [] : An Emerging Global Fuel wwwmethanolorg/energy/fuel/- Emerging-Global-Fuelaspx