Enhancing Flexibility and Transient Capability of the Diesel Engine System Simulation Zoran Filipi Dennis Assanis Dohoy Jung George Delagrammatikas Jennifer Liedtke David Reyes Doug Rosenbaum Alejandro Sales ARC Conference May 19-20, 1998 Ann Arbor, Michigan
ACKNOWLEDGEMENTS National Automotive Center (NAC) located within the US Army TARDEC for technical and financial support. The University of Wisconsin team has contributed the MATLAB vehicle template and component models for the drivetrain sub-system. Guoqing Zhang and Xiaoliu Liu for initial contributions to implementation of the V12 diesel engine simulation in Matlab-SIMULINK.
OUTLINE Introduction: - The need for enhanced flexibilty and transient capability of the diesel engine simulation in the context of Powertrain System modeling. Thermal network modeling of engine heat rejection. Assessment of the potential of the Low Heat Rejection (LHR) tank engine. Virtual diesel engine for the HEV. Scaling of the complete diesel engine system for optimization studies.
Flexibility Matlab-SIMULINK environment allows easy reconfiguration of the engine and powertrain system, e.g. variation of the number of cylinders, configuration of the driveline (4x2, 4x4, 6x6...). Variation of the number of cylinders or cylinder size requires resizing of external diesel engine system components. Turbomachinery is modeled using digitized maps, hence for every variation of engine size new set of maps is needed. Control devices, such as wastegates also need to be scaled.
Need to Enhance Transient Capability Rapid changes of engine speed and load initiate very dramatic thermal transients Combustion chamber thermal condition affects volumetric efficiency, heat rejection, combustion and friction in a diesel engine Transient heat transfer model needed to enhance simulations ability to predict system response and vehicle performance Enhanced heat transfer model essential for evaluation of alternative designs, such as Low Heat Rejection (LHR) engines
Thermal Network Modeling - MOTIVATION Wall temperature variations during engine speed and load transient 510 Piston Surface Temp. (K) 505 500 495 490 485 480 Cyclic fluctuations of wall temperatures 475 0 5 10 15 20 25 30 Time (s)
Thermal Behavior of the LHR Engine Cyclic surface temperature variations of conventional metal engines : 5 to 15 K 2000 Cyclic surface temperature variations of LHR engines: 100 to 150 K (Zirconia Coating) Larger temperature swing of LHR engine requires the capability for transient surface temperature prediction. Temperature (K) 1500 1000 500 Mean Gas ---- Piston Surface Conventional Eng. LHR Eng. (1.0 mm Coating) 0 0 100 200 300 400 500 600 700 Crank Angle (deg)
Modeling Issues Need fidelity and computational efficiency Finite Element Methods - High fidelity - Computationally intensive - Need to generate meshes for every new design Thermal Network model provides a good compromise: - High fidelity of global component temperature predictions - much less computational effort than FEA methods
Lumped Capacitance Method Sub-system: cylinder head, piston, liner, oil reservoir, coolant in head and block. Head Water Jacket Head Heat Exchanger Water Jacket Combustion chamber walls are divided into 8 sublayers based on Biot Number criterion: Bi hlc = <<1 k Cylinder Gas Piston Liner Block Heat Exchanger h : convection coef. Lc : characteristic length k : conductivity Oil Reservoir Oil Heat Exchanger
Heat flux term arc LUMPED CAPACITANCE METHOD (Mathematical Formulation) Conservation of Energy cond. conv. rad. T T source sin k p p + Q Q mc p source sin k = R j j p ij, i p Thermal Resistances R = L ka i v, i T i p + 1 T t T :Nodal Temp. R : Thermal Resistance Q : Heat Source or Sink m : mass Cv : Const. Vol. Specific Heat i, j : node p, p+1 : Time step t : Time Step Size (Axial Conduction) i p Capacitance term r r R = ln( 2 / 1 ) 2πHk R = 1 has (Radial Conduction) (Convection) L : Distance between Nodes k : Conductivity A : Cross Sectional Area r1, r2 : Inner and Outer Radii H : Cylinder Height h : Convection Coef. As : Surface Area
Virtual Tank Engine Air Exhaust gas Exhaust gas Air Hypothetical V12 Diesel Engine C T T C 4-Stroke DI Diesel 2 Turbochargers INTER- COOLER INTER- COOLER 2 Intercoolers Bore = 6.25 in (15.9 cm) I M E M E M I M Stroke = 6.25 in (15.9 cm) CR = 15 FUEL SYSTEM V12 ENGINE Predicted Power: 1440 HP@2100 rpm Ẇ
Engine System in SIMULINK PowerSim I/M 1 E/M 1 Heat Trans. Model I/M 2 Cylinders E/M 2
The Effect of Wall Insulation on System Steady-State and Transient Performance Steady-state performance at full load Acceleration from stand still with 100% driver s demand after engine has been warmed up Three virtual engine versions: - Conventional Engine - LHR Engine (0.5 mm Zirconia Coating) - LHR Engine (1.0 mm Zirconia Coating)
Engine Performance Comparison - Conventional vs. LHR Engine 6000 2000 5500 Torque (Nm) 5000 4500 4000 3500 1500 1000 Power (kw) 3000 500 Conventional Eng. LHR Eng. (0.5 mm Coating) 2500 LHR Eng. (1.0 mm Coating) Cummins Eng. 2000 0 800 1000 1200 1400 1600 1800 2000 2200 Engine Speed (rpm)
VEHICLE ACCELERATION Vehicle Acceleration from stand still for 40 seconds (Conventional Engine) Rotational Speed (rad/s) 250 Engine 200 150 Transmission In Transmission Out 100 50 Drive Shaft 0 0 5 10 15 20 25 30 35 40 Time (s) Torque (Nm) Shaft 30000 25000 20000 15000 10000 5000 Torque Converter Out Engine Sprocket 0 0 5 10 15 20 25 30 35 40 Time (s)
Transient Temperature Variations Piston Surface Temp. (K) 750 700 650 600 550 500 LHR Engine (1.0 mm Coating) Conventional Engine Exhaust Gas Temperature (K) 880 870 860 850 840 830 820 810 LHR Engine (1.0 mm Coating) Conventional Engine 450 0 5 10 15 20 25 30 Time (s) 800 0 5 10 15 20 25 30 35 40 Time (s)
Boost Pressure Histories - Conventional vs. LHR 400 Boost Pressure (kpa) 350 300 250 200 LHR Engine (1.0 mm Coating) Conventional Engine 150 100 0 5 10 15 20 25 30 35 40 Time (s)
System Response Fuel control based on manifold pressure Speed (Mile/h) Vehicle Comparison of Vehicle Speeds 60 50 40 30 20 LHR Engine 10 Conventional Engine Cylinder Equivalence Ratio (-) 0.7 0.6 0.5 0.4 0.3 0.2 0.1 F/A Equivalence Ratio in the LHR Cylinder LHR Engine 0 0 5 10 15 20 25 30 35 40 Time (s) 0.0 0 5 10 15 20 25 30 Time (s)
Enhancing Flexibility for Integration with Optimization Codes Engine system needs to be simulated within the range, e.g. 1.0 to 1.9 liter displacement. External components have to scaled accordingly, including turbomachinery. Continuous variation of size required throughout the range
I M arc ATMOSPHERE C INTER- COOLER FUEL SYSTEM T E M 4CYL ENGINE 2.6 2.4 2.2 2 1.8 1.6 1.4 1.2 1 Turbocharged Diesel Engine for the HEV WG Pressure Ratio Baseline engine : VW 1.9 L TDI Wide engine speed range requires boost pressure control - wastegate Simulated engine operating line Mass flow rate 0 5 10 15
Scaling of Engine Geometry Express the following as a function of Bore: - Stroke - Connecting rod length - Valve/port diameters and maximum valve lifts - Manifold volumes Assume scaled engine will have same S/B ratio as the baseline engine Find new Bore as a function of new displacement. i.e: B new B V old displ new = ( ) S Π I old 4 _ 1/ 3 cylinders Calculate new engine geometry as a function of B new
Wastegate Modeling Spring P atm... Wastegate valve dynamics m z+ b z+ Sz = ( p p ) A + ( p p ) A F exman back v cntrl atm dfgm prel m, b, S, A v, A dfgm, F prel - design parameters Pressures at every instant supplied by the engine simulation Diaphragm P BACK P EXMAN P control (boost) TURBUNE INLET SIDE Needs to be scaled along with engine geometry A v new /A v old = A dfgmnew /A dfgmold =V new /V old Spring stiffness and F prel scale linearly with A dfgm new
Dimensional Analysis Assuming Overview of Turbomachinery Scaling Methodology Based on non-dimensional representation of Compressor and Turbine Characteristics hence Scaling factor Scaleing of characteristics m p 01. same. RT D ND m D 01 2 T0 ND p02 m, η, = f(,,, γ) T RT p µ D = 01 01 ND 1 1 2 2.. 1 m2 1 2 D2 2 V V = ; η = η displ _ new displ _ old. 2 m. = α m 1 = α 1 2 D D 1 2 01 2.6 2.4 2.2 2 1.8 1.6 1.4 1.2 N2 1 = = N α 1 1 negligable same N max1 > N max 1.9 0 5 10 15
Simulated Intake and Exhaust Manifold Pressure for the Range of Turbocharged Engines 2 2.2 Intake Manifold Pressure (bar) 1.8 1.6 1.4 1.2 Displacement=1.9L Displacement=1.8L Displacement=1.7L Displacement=1.6L Displacement=1.5L Displacement=1.4L Displacement=1.3L Displacement=1.2L Displacement=1.1L Displacement=1.0L Exhaust Manifold Pressure (bar) 2 1.8 1.6 1.4 1.2 Displacement=1.9L Displacement=1.8L Displacement=1.7L Displacement=1.6L Displacement=1.5L Displacement=1.4L Displacement=1.3L Displacement=1.2L Displacement=1.1L Displacement=1.0L 1 1000 2000 3000 4000 Engine Speed (RPM) 1 1000 2000 3000 4000 Engine Speed (RPM)
Simulated Power Output and BSFC for the Range of Turbocharged Engines 60 270 Power (kw) 50 40 30 20 10 0 Displacement=1.9L Displacement=1.8L Displacement=1.7L Displacement=1.6L Displacement=1.5L Displacement=1.4L Displacement=1.3L Displacement=1.2L Displacement=1.1L Displacement=1.0L 1000 2000 3000 4000 Engine Speed (RPM) BSFC (g/kw-hr) 260 250 240 230 220 210 200 Displacement=1.9L Displacement=1.8L Displacement=1.7L Displacement=1.6L Displacement=1.5L Displacement=1.4L Displacement=1.3L Displacement=1.2L Displacement=1.1L Displacement=1.0L 1000 2000 3000 4000 Engine Speed (RPM)
Summary Thermal network modeling allows prediction of the effect of the variation of engine component wall temperatures on system response and vehicle performance Thermal network model critical for evaluation of the LHR concept for tank propulsion Lumped capacitance model provides fidelity at low cost Turbomachinery scaling methodology enhances the flexibility of the system simulation and allows continuous variations of engine size in optimization studies.
Future Challenges Extend the thermal network model to include engine external components, e.g. manifolds. Investigate engine transients under extreme conditions, i.e.: - cold start and engine acceleration at very low temperatures - system response at very high ambient temperatures - high altitude operation Develop techniques for modeling variable geometry turbines/compressors. Investigate the effect of alternative turbocharging techniques, e.g. sequential turbocharging, supercharging + turbocharging etc.