PAYET-BURIN Thomas (Alten) Direction de la Qualité et de l Ingénierie
Agenda Introduction Hydraulic studies Overview of the architecture of a hybrid cooling system Hydraulic correlation Architecture choices Filling studies Collaboration with FIVES Correlation - Diesel Engine Projection Gasoline Plug-In Hybrid Engine Integration of a cooling system in a heat distribution model Conclusion 2
Introduction Context: Development of a new modular platform: a plug-in petrol hybrid Available in 2 and 4 wheel drive versions in 2019 Develop a low carbon solution for each body style, while maintaining the highest standards in performance and cabin space Motivations for cooling system modeling: Determine the flow rate in each branch of the circuit to ensure that the minimum flow rate in each branch is respected Optimization of the circuit in the early phase of the V-Cycle No need of a prototype engine to optimize the cooling system Goals: Ability to make some study on a cooling system using only the CAD and predictive simulation tools (1D / 3D) The error should be less than 50 mbar on the pressure and less than 5% on the volumetric flow rate 3
Agenda Introduction Hydraulic studies Overview of the architecture of a hybrid cooling system Hydraulic correlation Architecture choices Filling studies Collaboration with FIVES Correlation - Diesel Engine Projection Gasoline Plug-In Hybrid Engine Integration of a cooling system in a heat distribution model Conclusion 4
Overview of the architecture of a hybrid cooling system 5
Overview of the architecture of a hybrid cooling system Different loops are existing in a hybrid cooling system High Temperature coolant circuit loop HT Low Temperature coolant circuit loop LT Very Low Temperature coolant circuit loop VLT CAC coolant circuit loop Different loops could be combined for cost and packaging reasons 6
Gasoline plug-in hybrid engine LT & VLT cooling system architecture LT and VLT loops are combined 2 operating modes: 1. Short loop: i. Radiator Pump Inverter 1 Inverter 2 Electrical machine ii. Chiller Pump Battery Charger Battery is cooled down by the chiller 7
Gasoline plug-in hybrid engine LT & VLT cooling system architecture LT and VLT loops are combined 2 operating modes: 1. Short loop: i. Radiator Pump Inverter 1 Inverter 2 Electrical machine ii. Chiller Pump Battery Charger Battery is cooled down by the chiller 2. Long loop: Pump Inverter 1 Inverter 2 Electrical machine Radiator Chiller Battery Battery is cooled down by the radiator Or Battery is heated by the electrical machine and inverters 8
Model building GT-Ise Final model GEM3D - Conversion to 1D element SpaceClaim - Volume Extract SpaceClaim - Conversion of the CAD data 9
Model building SpaceClaim Simple steps to extract the volume Extract the exact geometry in order to be as predictive as possible (prediction of the pressure losses) GT-Ise Final model GEM3D - Conversion to 1D element SpaceClaim - Volume Extract SpaceClaim - Conversion of the CAD data 10
Model building GEM3D Inlet / outlet port automatically detected Automatic export to 1D element Representation of the geometry as close as possible to the reality GT-Ise Final model GEM3D - Conversion to 1D element SpaceClaim - Volume Extract SpaceClaim - Conversion of the CAD data 11
Model building GT-Ise The components should be added to the model: HeatAddition for heat exchangers, inverter, charger and battery Pump for electrical pumps FluidReservoir for degas bottle Test conditions are defined Temperature Pressure Fluid composition GT-Ise Final model GEM3D - Conversion to 1D element SpaceClaim - Volume Extract SpaceClaim - Conversion of the CAD data 12
Pressure loss (mbar) Hydraulic correlation - Piping 140 Pressure loss of the pipe for the LT and VLT loops Comparison between Test and Simulation Test Simulation 120 100 80 60 40 22 meters of pipes 20 0 Chiller / Battery Battery / Pump Pump / Charger Charger / Chiller Pump / Inverter Inverter / Electrical Machine Electrical Machine / Radiator Radiator / Pump Good prediction of the pipe pressure loss using SpaceClaim / GEM3D Average error = 7,1 mbar Max error = 20,6 mbar Below the maximal authorized error of 50 mbar 13
Pressure (bar) Pressure (bar) Hydraulic correlation Complete Circuit 2.00 Short Loop 1-25 C Test Simulation P11 1.50 1.00 0.50 0.00 P2 P3 P4 P13 P14 P15 P20 P1 P15 P4 P3 P20 P2 P1 P12 2.50 Short Loop 2-25 C Test Simulation P14 P10 2.00 1.50 1.00 P13 P7 P8 P9 0.50 0.00 P8 P9 P10 P11 P12 P5 P6 P7 P6 P5 14
Pressure (bar) Hydraulic correlation Complete Circuit 2.00 Long Loop - 25 C Test Simulation 1.50 1.00 0.50 0.00 P7 P8 P9 P10 P11 P12 P5 P6 P13 P14 P15 P20 P1 P2 P3 P4 15
Pressure (bar) Pressure (bar) Pressure (bar) Hydraulic correlation Complete Circuit 2.00 Short Loop 1-25 C Test Simulation 2.00 Long Loop - 25 C Test Simulation 1.50 1.00 0.50 1.50 1.00 0.00 2.50 2.00 1.50 1.00 0.50 0.00 P2 P3 P4 P13 P14 P15 P20 P1 Short Loop 2-25 C Test Simulation P8 P9 P10 P11 P12 P5 P6 P7 0.50 0.00 P7 P8 P9 P10 P11 P12 P5 P6 P13 P14 P15 P20 P1 P2 P3 P4 Average error = 20 mbar Max error = 47 mbar Volumetric flow rate error = 0,17 L/min (~2% error) The results are quite satisfactory since only one point was given for the pressure loss versus flow rate curve for each component and the maximal error is below the target of 50 mbar with only 3 modifications to the baseline model (calibration of 3 orifice diameter) 16
Architecture choices Architecture choices are done using GT-Suite Component choices Electrical pump Choice of combination of loops Those studies are done in order to optimize the architecture under packaging constraint while keeping in mind that a minimum flow rate should be ensure in each branch S2 E S1 OBC_DCDC Onduleur DCDC Machine Mel Elec etar - Chiller Batterie + EE Onduleur 17
Agenda Introduction Hydraulic studies Overview of the architecture of a hybrid cooling system Hydraulic correlation Architecture choices Filling studies Collaboration with FIVES Correlation - Diesel Engine Projection Gasoline Plug-In Hybrid Engine Integration of a cooling system in a heat distribution model Conclusion 18
Filling studies Motivations for filling studies: Large volume (~14L) Only one filling point Is it doable? Goals: Determine the time needed during the evacuation process to have the pressure lower than a given pressure Determine the time needed to fill the cooling circuit depending initial conditions Work in progress: Collaboration with Correlation on diesel engine Projection on gasoline plug-in hybrid engine HT + LT loops VLT loop 19
Collaboration with FIVES Filling and evacuation machine model Hydraulic model Evacuation model Filling model 20
Correlation on the diesel engine Evacuation Initial conditions: Test: air & coolant / 1 bar / 25 C Simulation: dry air / 1 bar / 25 C Comments: Good correlation up to 10 seconds 2L of coolant were initially present in the circuit during the test. For this reason the pressure can t go lower than 34 mbar which is the saturation vapor pressure of coolant. For the moment, the model can t be used to determine the time needed to go lower than a given pressure 21
Correlation on the diesel engine - Filling Initial conditions: Test: air & coolant / 34 mbar / 25 C Simulation: dry air / 34 mbar / 25 C Comments: Good correlation on the coolant volume : error of 0,2L at 14 seconds The model can be used to determine the time needed to fill a circuit since the error is 0,5 seconds Poor correlation on the pressure Is it link to the fact that there was coolant initially in the circuit? 22
Projection on Gasoline Plug-In Hybrid Engine loops Initial conditions: dry air / 11 mbar / 25 C Sweep of maximal authorized pressure in the filling machine Very Low Temperature loop (~3,5L): High Temperature and Low Temperature loops (~14L): 23
Agenda Introduction Hydraulic studies Overview of the architecture of a hybrid cooling system Hydraulic correlation Architecture choices Filling studies Collaboration with FIVES Correlation - Diesel Engine Projection Plug-In Hybrid Engine Integration of a cooling system in a heat distribution model Conclusion 24
Integration of a cooling system in a heat distribution model Perimeter: The heat distribution model is constituted of: The oil circuit The coolant circuit The engine block (cylinder head, block, crankcase, ) The exhaust manifold Goals: Optimize the thermal hydraulic architecture Optimize the engine mechanical efficiency Ensure the thermo-mechanical resistance of the engine block Ensure the thermal resistance of the lubricant Input data: CAD Coolant circuit Oil circuit Combustion Air path circuit Output data: Oil, coolant and material temperature Thermal energy balance 25
Agenda Introduction Hydraulic studies Overview of the architecture of a hybrid cooling system Hydraulic correlation Architecture choices Filling studies Collaboration with FIVES Correlation - Diesel Engine Projection Plug-In Hybrid Engine Integration of a cooling system in a heat distribution model Conclusion 26
Conclusion Hydraulic studies: Good correlation of pipes pressure losses thanks to SpaceClaim and GEM3D Optimization and architecture choices thanks to the prediction of GT-Suite Filling studies: Results are encouraging for the evacuation and filling process For the evacuation process A new correlation should be made on a new engine with dry air as initial conditions For the filling process: Good correlation for the filling time Poor correlation on the pressure Work in progress with FIVES Correlation on a new engine with dry air as initial conditions 27
Thank you for your attention Questions? 28