OpEneR Optimal Energy Consumption and Recovery based on system network

OpEneR Optimal Energy Consumption and Recovery based on system network Deliverable 2.3: Report documenting vehicle specification, topology and simulation results. Also running vehicle delivery Version 0.3 Dissemination PU Project Coordination PCA Due Date 30.08.2012 Version Date 20.08.2012 7th Framework Programme GC.ICT.2011.6-8 PPP GC: ICT for fully electric vehicles Architectures for Energy, Communication and Thermal Management Grant Agreement No. 285526 STREP Small or medium-scale focused research project www.fp7-opener.eu

Authors RB Jochen Steinmann Kosmas Knödler PCA AVL Vincent Abadie Sylvain Laversanne Stephen Jones Arno Huss Rolf Albrecht Emre Kural Project Coordinator Dr. Kosmas Knödler Robert Bosch GmbH Chassis Systems Control Robert-Bosch-Allee 1 74232 Abstatt Germany Phone: +49 7062 911 2044 Fax: +49 711 811 5184887 Email: kosmas.knoedler@de.bosch.com Copyright 2011: the OpEneR Consortium Page 2 of 19

Version control Version Author Date Description 0.1 Sylvain Laversanne, Vincent Abadie 22.07.2012 Generation of Report 0.2 Emre Kural 10.08.2012 Simulation Results is added 0.3 Jochen Steinmann, Kosmas Knödler 20.08.2012 Final version Page 3 of 19

Table of Contents Version control...3 Table of Contents...4 Figures...5 Summary...6 1 Introduction...7 2 Specifications and Targets for Base Concept Vehicle...8 2.1 Base prototype concept...8 2.2 Specification of Base Concept Vehicle...9 2.3 Final specifications after first design freeze...10 3 Topology...11 3.1 Mechanical implementation...11 3.2 Power electronics topology and high voltage cables...13 4 First Simulation Results with ESP hev and Simulation Environment...14 4.1 Concept Simulation...14 4.2 Traffic scenarios simulations...15 4.3 ESP hev Simulation...17 Page 4 of 19

Figures Figure 1: The 2 OpEneR prototypes demonstration during first project review...7 Figure 2: Base Concept Vehicle...8 Figure 3: 3D view of the OpEneR Powertrain...11 Figure 4: Available space under driver & passenger seats...12 Figure 5: Battery package topology...12 Figure 6: Bottom view, topology of the high power & voltage cables...13 Figure 7: powertrain implementation...13 Figure 8: Definition of the required rated and peak torque level of the e-machine...14 Figure 9: AVL Cruise model for Concept Simulation...15 Figure 10: Traffic Jam Scenario in OpEneR simulation tool-chain...16 Figure 11: Generic Route network...16 Figure 12: Simulation results comparing the velocity profile, SOC, energy consumption, travel time over 2 alternative road: PATH1, shortest path and PATH2, recommended direction by OpEneR function...16 Figure 13: ESP module integrated into CarMaker4SL...17 Figure 14: AVL Cruse model customized for CarMaker co-simulation and ESP hev application...17 Figure 15: Constant braking on normal road surface...18 Figure 16: µ-split manoeuvre (µleft = 1; µright = 0.2)...18 Figure 17: µ-split manoeuvre (µleft = 1; µright = 0.2)...19 Figure 18: Constant braking on µ split road...19 Page 5 of 19

Summary In order to integrate new OpEneR technologies and energy manager, a new concept of FEV has been developed. 2 prototypes were designed, not only for demonstration purpose, but also as a working base for measurement, real time implementation and testing. The prototype specification was adapted so the vehicle could run in most condition from low temperatures in winter testing to high temperatures in Spain for driving on CTAG intelligent corridor. The protective design allows the addition of new elements such as innovative braking systems, surround sensing systems, SatNav navigation system and new HMI. The specific architecture of the OpEneR prototype, being Full Electric and 4WD with 2 e-machines, imposed a very specific topology, with powertrain elements spread from the front to the rear of the vehicle. After the first 12 months of the project, the 2 prototypes were demonstrated during the first project review. Those prototypes are the running proof of the concept, with high driveability and maximum regenerative braking. Those prototypes are now ready to integrate the next OpEneR levels. Page 6 of 19

1 Introduction Deliverable 2.3 describes one of the outcomes of WP2, that is to say the design and delivery of the 2 OpEneR prototypes after the first 15 months of the project. This WP2 mainly deals with: Vehicle specification and base prototype delivery Concept of system topology on vehicle level including all new subsystems Design, dimensioning and simulation of the vehicle concept (dependent on functional and geometric including specification and packaging) Definition of metrics to quantify the effectiveness of strategies and definition of a reference vehicle Control and operation strategies with respect to safety, comfort and efficiency according to OpEneR Level 1, Level 2 then Level 3. Overall energy manager with overall operation strategies Those prototypes have been delivered to the partners and have been demonstrated during the first project review (see Figure 1). Prototype OpEneR 1 has been delivered in March 2012 to BOSCH after a winter test session. OpEneR prototypes being far from standard serial production vehicles, this early delivery allowed us to make the necessary validations (Thermal management, Powertrain, Acoustic, etc.). With anticipation new OpEneR braking components were integrated and calibration started. Prototype OpEneR 2 was built just after OpEneR 1 using the first feedback from its validation, so the workshop phase would be shorter and the validation phase reduced. The prototype was delivered in May 2012 to CTAG in order to integrate the OpEneR HMI. Figure 1: The 2 OpEneR prototypes demonstration during first project review Page 7 of 19

2 Specifications and Targets for Base Concept Vehicle 2.1 Base prototype concept The base prototype vehicle for the project is a Peugeot 3008 FEV1 e-4wd2 vehicle provided by PSA. This vehicle is currently not under serial production. The prototype will be adapted from the serial production 3008 Hymotion4. As shown in Figure 2, the e-4wd diesel hybrid vehicle will be transformed into a Full EV with front and rear electric axles allowing maximum regenerative braking. Figure 2: Base Concept Vehicle The Base Concept Vehicle facts can be summarized as follows: Baseline vehicle is a 3008 Hymotion transformed into Full electric 4WD vehicle The front conventional engine of the 3008 Hymotion is replace by e-machine The rear serial production e-machine should be kept New battery package for Base Concept Vehicle should be adapted ESP / Master vacuum cylinder should be removed to fit in ESP hev / ibooster Accessories should be adapted, e.g. Heating, cooling, charging device, HMI This new vehicle concept with rear and front electrified axles is specially designed for the OpEneR project : Thanks to this e-4wd FEV architecture, energy recovery can be optimised up to more than 0.4g electric deceleration with complete safety, even on slippery road Two axles allow very high torque at low speed improving agility and adaptation for city driving hence increasing of attractiveness One axle can be decoupled at higher speed to allow very low losses from only one of the two e-machines. It shows the best performance, e.g. with the new generations of regenerative brake algorithms 1 Full Electric Vehicle 2 Electric 4-Wheel-Drive Page 8 of 19

2.2 Specification of Base Concept Vehicle The OpEneR Base Concept Vehicle targets could be summarized as follows: Minimum electric range should be 150 km in urban driving Maximum reachable Vehicle velocity should be 140 km/h (for flat Highway driving) City driveability should be appropriate: Minimum 0.3g acceleration at low speed (~3 m/s²) Specific power should be at minimum 40 kw / t Expected prototype lifetime is 4 years / 150000 km mileage Same e-machine size must be used on both axles to avoid additional costs Maximum reuse of serial production components The OpEneR prototypes being Full Electric Vehicles, one major constraint from specification point of view is the environment temperature, especially for the battery system which runs at moderate temperature levels. Furthermore heating and cooling the cockpit of a FEV is critical and leads to major consumption of energy, therefore it must be taken in account by the overall energy manager. The operating temperature range for the prototypes is defined as follows: D C B A B C D -30 C -20 C -5 C 30 C 35 C 40 C A. Nominal between -5 C and 30 C. The prototype should be running at maximum performance B. Rugged condition between -20 C and -5 C and between 30 C and 35 C. Maximum performance should be available for a limited time. Continuous electric power is 2/3 of max power C. Exceptional conditions between -30 C and -20 C and between 35 C and 40 C. 2/3 of max power will be available for a limited time, ½ max power for lowest temperatures. Continuous electric power is 1/3 of max power D. Out of range conditions is below -30 C and above 40 C. The prototype shouldn t run in those conditions. Page 9 of 19

2.3 Final specifications after first design freeze 2x50kw (mechanical power) e-machines 40kwh battery package 4x125kg package in serial connection FeSO 4 technology 36,8 kwh usable capacity (200km range) 110kw discharge and charge (depending on temp.) Thermal management 2 radiators for inverters/e-machines and battery Water heating via 2x 6kw CTP (cockpit & battery) Electric A/C compressor with double circuit (battery & cockpit) Body & chassis design New Body design for battery integration General behavior similar to 3008 hybrid4 (+250kg) Body Stiffness or better same as serial production Page 10 of 19

3 Topology The topology of this new vehicle concept is mainly driven by space constraints. In order to fit a 40 kwh battery in the Peugeot 3008, it is necessary to divide the battery package in order to make the mass even through the vehicle structure. From that mechanic implementation, the power electric and network topology follow. 3.1 Mechanical implementation Battery Package Front & Rear inverters & DCDC Rear e-machine Front e-machine Figure 3: 3D view of the OpEneR Powertrain The battery package is divided in 4 packages (see Figure 3). Each package weight is 125kg. In order to keep a good behaviour with a heavier vehicle, the centre of gravity needs to stay as low as possible. A higher centre of gravity would mean that the prototype would be more susceptible to roll, which could lead to a worse drivability and to one extent would become a safety issue. Other constraints must be taken in account, such as crash and yaw inertia, which drive the design to maximize the mass of the vehicle near its centre of gravity. A low yaw inertia allows faster steering response, and better drivability. Furthermore, keeping the mass in the centre of the vehicle keeps the battery packs away from external object intrusion in case of crash. One last issue is to let the battery packages vent in a very rare case of failure. This venting is necessary for safety, and the fumes are toxic in a closed area. So the implementation of the battery needs to be such that the packages are outside the main body but still protected from external mechanical impacts or water spray. Page 11 of 19

Figure 4: Available space under driver & passenger seats Thanks to a high seating position in the original Peugeot 3008, with a new floor design and thinner seats (see Figure 4), some space could be found in order to integrate the new battery packages. Figure 5: Battery package topology Three battery packages were then integrated under a new designed floor and the last package in the front area within the remaining space (see Figure 5). This position allows the centre of gravity to be as low as possible (position is 2 cm lower than serial production 3008 Hybrid4), and the overall mass to be centred with low yaw inertia. Due to the lack of space in the front compartment, the inverter for the front e-machine has been implemented in the trunk. Page 12 of 19

3.2 Power electronics topology and high voltage cables Since the powertrain is composed of two e-machines and 4 battery packs, high power cables go across the vehicle from the rear to the front. Furthermore, power cables (50mm 2, shielded) high rigidity implies a very early design of the cable layout within the first design of the prototype (see Figure 6 and Figure 7) The 4 battery packs are connected in serial connection The inverter for the front e-machine is located in the trunk of the vehicle All high voltage cables are regrouped under the floor in a duct, in the centre of the vehicle for maximum safety and are separated from the interior for maximum Electro-magnetic compatibility R-inverter F-inverter PACK2 R-e-mach PACK1 BATTERIES ARE IN SERIAL CONNECTION PACK3 Figure 6: Bottom view, topology of the high power & voltage cables Inverters Rear e-machine Battery packs High voltage cables duct Front e-machine Figure 7: powertrain implementation Page 13 of 19

4 First Simulation Results with ESP hev and Simulation Environment The details of the developed simulation environment and simulation models were already discussed in several Deliverable Reports (D4.1, D2.2) with different aspects. In this deliverable report, the summary of some of the results obtained with several simulation models will be listed here. 4.1 Concept Simulation Within WP2 the initial concept vehicle simulation is represented by Task 2.3 - Design, dimensioning and simulation of vehicle concept. This initial simulation is executed to confirm the final vehicle, powertrain and electrical specification of the OpEneR Base Concept Vehicle. It is done based on EXCEL calculations to initially specify the electric components (e-machine & Battery) based on predefined vehicle boundaries and requirements, followed by 1D rotational AVL CRUISE simulations based on a CRUISE simulation model validated towards the EXCEL calculation results for NEDC and performance. The validated CRUISE model is finally used to evaluate significant quantification parameters in terms of a recuperation potential analysis as well as the optimization of the appropriate e-traction torque distribution factors to maximize the overall system efficiency. In Figure 8, a representative set of simulation results for different drive cycle are summarized. This study concludes with the definition of the dimensioning of powertrain components with respect to the defined OpEneR vehicle targets. Further details of this study can be found in Deliverable Report 2.3. emotor Torque (Nm) 400 300 200 100-100 -200-300 -400 Driving Resistance Torque located @ emotor vs. emotor Speed - OpEneR EV 0 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 emotor Speed (rpm) Gradient of 0% Gradient of 10% Gradient of 15% Gradient of 20% Gradient of 25% Gradient of 30% Gradient of 40% Gradient 0f 0% 1m/s² Gradient of 0% 2m/s² Gradient of 0% 3m/s² Gradient of 0% 5m/s² Artemis Highway Artemis Road Artemis Urban Auto Motor Sport NYCC SC03 US06 NEDC FTP 75 Highway TARGET: Acceleration TARGET: Max. Velocity emotor Peak emotor Continuous Figure 8: Definition of the required rated and peak torque level of the e-machine Page 14 of 19

Figure 9: AVL Cruise model for Concept Simulation 4.2 Traffic scenarios simulations The developed simulation environment (see Figure 9) enables to represent several realistic traffic situations and enables to interaction of traffic objects. Such a special scenario is represented in Figure 10. Basically, this special simulation scenario tests a developed OpEneR functionality that recommends to the driver an optimal route selection considering the vehicle energy consumption characteristics, real-time traffic situation around the road network and road topology. As seen in Figure 11, until the destination there are several route alternatives with different traffic load condition. The optimal route planning functions considers the energy to be consumed on the shortest path with stop-and-go traffic and compares to it to longer route selection. Finally it suggests to the driver the selection of the longer path, consequently results with lower energy consumption and less travel time (see Figure 12). Page 15 of 19

Figure 10: Traffic Jam Scenario in OpEneR simulation tool-chain Fast Average speed traffic Slow STOP g NO g KL g LM g MN K L M N O g KF g GL g HM g NI g OJ g FG g GH g HI g IJ F G H I J g AF g BG g CH g DI g EJ START PATH 1 PATH 2 g AB g BC g A B CD g C D DE E Velocity in km/h 80 60 A B C 40 20 D Figure 11: Generic Route network Vehicle Velocities E I Path 1 Path 2 J O N N PATH 1 PATH 2 SOC in % 0 0 0.5 1 1.5 2 2.5 3 3.5 Distance [km] 100 Battery State of Charge Path 1 Path 2 99.5 99 ΔE = -10.3% E = 1.98 kwh E = 1.79 kwh T = 454s ΔT = 182s T = 272s 98.5 0 0.5 1 1.5 2 2.5 3 3.5 Distance in km Figure 12: Simulation results comparing the velocity profile, SOC, energy consumption, travel time over 2 alternative road: PATH1, shortest path and PATH2, recommended direction by OpEneR function Page 16 of 19

4.3 ESP hev Simulation Another specific simulation test-case is about the integration of the Bosch ESP hev control module into the simulation environment. Although this model is not yet calibrated for OpEneR vehicle in the simulation environment, the integration and interface definition has already been realized. The successful integration of the ESP hev module into the whole simulation environment enables to make some functional and qualitative simulation tests. Quantitative testing will only be realized once the calibrated ESP hev module is available. In Figure 13 and Figure 14, the integrated ESP hev software to simulation tool-chain is seen. Sync _Out Sync _In Am bient.mis c CM _FIRST Sync _Out Sync _In Sync _In Sync _Out Sync _In Sync _Out VhclCtrl.Steering VhclCtrl.Steering VhclCtrl.Brake VhclCtrl.Brake PowerTrain.Misc VhclCtrl.PT VhclCtrl.PT Ambient.MiscDrivMan.Out DrivMan.In Brake.IF.In Brake.IF.In Brake.IF.Out.Brk VehicleControl Brake.IF.Out.PT DrivMan Vehicle ESP Figure 13: ESP module integrated into CarMaker4SL Sync _In Sync _Out PowerTrain.Misc CM_LAST Figure 14: AVL Cruse model customized for CarMaker co-simulation and ESP hev application Page 17 of 19

ESP hev module has been tested in simulation environment to check the functionalities over different road condition. The results presented in this report show how the operation of recuperation is adapted to different road surface conditions. In Figure 15, the model is tested under normal road surface and braking maneuver is realized just after an acceleration phase. During braking maneuver, ESP hev enables the recuperation for almost whole deceleration phase. This recuperation phase is represented by the torque blending flag (flag=1 shows the recuperation area).the increase of the Battery SOC can also be observed. Yaw Rate [rad/s] 0.10 60 0.05 40 0.00 20-0.05 0 Brake Pedal Position [%] Torque Blending Flag 1.0 0.5 0.0 88.00 87.75 87.50 SOC [%] Actual EM Regen Torque [Nm] 0-100 -200-300 0 10 20 30 40 50 Time [s] Figure 15: Constant braking on normal road surface 60 40 20 0 Car Speed [m/s] Second simulation is performed when the surface condition is different for right and left side of the vehicle (µ left = 1; µ right = 0.2) during some part of the braking phase (see Figure 16 and Figure 17). Figure 16: µ-split manoeuvre (µleft = 1; µright = 0.2) Page 18 of 19

Figure 17: µ-split manoeuvre (µleft = 1; µright = 0.2) Torque blending flag, when the vehicle passes over the low µ part of the road prevents the braking with the recuperation. Once this part of the road is over, recuperation is allowed again by ESP hev (see Figure 18). Yaw Rate [rad/s] 0.10 60 0.05 40 0.00 20-0.05 0 Brake Pedal Position [%] Torque Blending Flag 1.0 0.5 0.0 88.00 87.75 87.50 SOC [%] Actual EM Regen Torque [Nm] 0-100 -200-300 0 10 20 30 40 50 Time [s] Figure 18: Constant braking on µ split road 60 40 20 0 Car Speed [m/s] Page 19 of 19