Real-world to Lab Robust measurement requirements for future vehicle powertrains

Real-world to Lab Robust measurement requirements for future vehicle powertrains Andrew Lewis, Edward Chappell, Richard Burke, Sam Akehurst, Simon Pickering University of Bath Simon Regitz, David R Rogers Kistler Instrumente AG

Agenda and Content Introduction Motivation and background Experimental set-up and testing Analysis and Results Future direction Summary Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Introduction Evolution of the powertrain is underway Legislation, Consumers and Manufacturers are continually seeking the following vehicle improvements: Better fuel economy Lower Emissions Higher Performance Lower Cost These demands are only going to intensify 4 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Challenge of Real Driving Emissions Implementation of WLTP and RDE in September 2017 to reduce the gap between Lab and Reality More rigorous test procedures Main target for RDE controls is NOx and PN emissions. Creates new development challenges considering the targets and boundary conditions Clear comparison between powertrain architectures is difficult to quantify 6 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Market drive towards electrification UK Government proposing minimum 50 mile EV range by 2040 for all new vehicle sales Increased complexity leads to increased difficulty for optimisation Knowledge of the energy flows is essential in order to optimise the: Hybrid operating strategy Efficiency of the components and control strategy Driveability 7 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Background Propulsion systems for hybrid vehicles can take different forms: Series de-coupled ICE from output drive Parallel ICE and EM can provide output drive Series/parallel combinations Combinations can become extremely complex combining the benefits of both series and parallel designs For each of these hybrid concepts the battery capacity defines the electrical energy availability and therefore re-charging from the grid determines the tailpipe CO 2. The propulsion system requires integration into the overall vehicle design at the initial concept phase. 8 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

The Test Vehicle BMW i8 plug in hybrid vehicle (model year 2016) 2-speed transmission 1.5L 3-cylinder Turbocharged SI engine 6-speed automatic transmission 96kW electric motor 7.1 kwh Battery Power electronics High voltage starter motor and battery charger 10 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

The Test Vehicle Vehicle is equipped with three driver-controlled modes: Electric only mode Standard comfort mode Sport mode 11 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Instrumentation Instrumentation designed to capture the major energy flows within the powertrain Vehicle was instrumented as follows: Kistler Roadyn system with torque measurement wheels Real time combustion pressure analysis with a Kistler KiBox system Vehicle Bus CAN data from the EMS (Engine Management System) 12 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Test cycles Road Features of RDE Urban Rural Highway Repeated in different driving modes and battery SOC Chassis Dyno 4-wheel drive chassis dynamometer facility at the University of Bath WLTC test cycles Repeated in different driving modes and battery SOC Test number Start Vehicle Driving Battery state of charge temperature Mode Start (%) End (%) ΔSoC (%) 1 Warm Comfort 85 46-39 2 Warm Comfort 32 40 +8 3 Cold Comfort 97 43-54 4 Warm Sports mode 43 82 +39 1 15 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Results and discussion Results section was classified into two parts: High level hybrid strategy over the different WLTC drive cycles, Highlighting the key differences depending on the battery state of charge and the driving modes. Detailed energy balance from both the chassis dynamometer and on-road driving Highlighting some of the analysis that can be conducted with the proposed level of instrumentation. 17 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Results and discussion Wheel Torques Hybrid strategy analysis Breakdown of tractive effort (Power and Energy) by axle for all repeated WLTC cycles In comfort mode rear axle energy usage proportional to SOC. Battery SOC depleting 18 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Results and discussion Wheel Torques Hybrid strategy analysis Analysis of the breakdown of axle power over each of the four phases of the WLTC cycle. The results show in all tests, the vehicle uses the rear axle considerably during high and extra high phases Little test-to-test variation in recuperation through the front axle in the extra high phase SOC and driving mode significantly effect the front/rear axle split. 19 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Speed [km/h] WPED [%] Charge Energy [kwh] Boost Energy [kwh] Results and discussion Powertrain Energy flows WLTC 2.0 1.8 1.6 1.4 1.2 Boost Front Axle Boost Rear Axle Total Boost Energy Estimated Boost and Charge Energies over a complete WLTC determined from driving condition, wheel and ICE power. 1.0 0.8 0.6 0.4 0.2 0.0 1.85kWh total electrical energy used to drive the vehicle 1.2 1.55kWh front axle Theoretic Kinetic Energy f rom Deceleration Brake Regeneration ICE Load Point Shif t Front Driv e Charging Total Charge Energy 1.0 0.8 0.6 0.3kWh rear axle 125.0 100.0 75.0 50.0 25.0 0.0 0 100 VSS WPED 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 0.4 0.2 0.0 100 80 60 40 20 0 1.1kWh of energy provided to the battery 0.7kWh KE recovery from decelerations 0.3kWh front axle charging 0.08kWh engine mounted e-machine Time [sec] 20 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Results and discussion Powertrain Energy flows WLTC Overall energy values of three different WLTC cycles The theoretical maximum kinetic energy recovery is similar for all cycles as this is defined by the vehicle speed trace of the WLTC Recuperation efficiency estimated to be in the region of 79%. Significant differences can be seen in the hybrid charging and boosting strategy 21 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Speed [km/h] WPED [%] Power [kw] Results and discussion Powertrain Energy flows On-road Ef f ectiv e Power HV Sy stem ( h electric driv e = 0.8) Ef f ectiv e Power ICE Power to Road (Wheels) Ef f ectiv e Power sum (ICE + HV Sy stem) 80.0 70.0 60.0 50.0 Torque blending situation wheel torque exceeds estimated powertrain torque 40.0 30.0 20.0 10.0 0.0-10.0 Electromechanical torque is estimated (no shaft measurement available) mechanical brake power -20.0-30.0-40.0-50.0-60.0-70.0 Difference between estimated effective power and wheel power is mechanical braking -80.0 100 100 75 50 VSS WPED 75 50 25 25 0 0 1891 1893 1895 1897 1899 1901 1903 1905 1907 1909 1911 1913 1915 1917 1919 1921 1923 1925 1927 1929 1931 Time [s] 22 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Results and discussion Powertrain Energy flows On-road measurements Energy fractions of the different operating states for RDE cycles On-road-data shows a bigger variation in comparison to the WLTC data Reduction in recuperation efficiency 23 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Speed [km/h] WPED [%] Power [kw] Results and discussion Powertrain Energy flows On-road measurements Ef f ectiv e Power HV Sy stem ( h electric driv e = 0.8) Ef f ectiv e Power ICE Power to Road (Wheels) Ef f ectiv e Power sum (ICE + HV Sy stem) 130.0 120.0 110.0 100.0 90.0 80.0 Boost Situation in sport mode, the powertrain is trimmed for maximum torque response 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0-10.0 During steady state driving in sport mode, the front axle is generating negative torque to pre-tension the powertrain 100 75 VSS WPED Boosting on fast throttle transient -20.0-30.0-40.0-50.0-60.0 100 75 The result is a rapid response on fast pedal transients as the pre-loaded ICE torque is released instantly 50 50 25 25 0 0 924 926 928 930 932 934 936 938 940 942 944 946 948 950 952 954 Time [s] 24 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Further work Addition of drive-shaft torque & BSG current measurements Addition of drive-shaft torque is beneficial to understand the power flux to the wheels All four drive-shafts have been instrumented BSG current measurement for direct determination of electric propulsion fraction on the rear axle Drive shaft torque telemetry Vehicle Bus 26 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Speed [km/h] WPED [%] Power [kw] Further work Powertrain energy flow the benefit of drive-shaft torque Torque Blending situation Harvested and lost energy can be precisely determined Excellent correlation between shaft & wheel power in normal driving conditions Addition of drive-shaft torque greatly improves analytic possibilities in regard to torque blending & torque vectoring 100 75 50 25 Regenerative Braking Shaft power VSS WPED Mechanical Braking Power electric driv e (Front Driv e shaf ts + BSG) Ef f ectiv e Power ICE Power to Road (Wheels) Power to Wheels (Shaf ts) Wheel power 50.0 45.0 40.0 35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0-5.0-10.0-15.0-20.0-25.0-30.0-35.0-40.0-45.0-50.0 Brake (mech. & regen.) 100 75 50 25 0 0 1443 1444 1444 1445 1445 1446 1446 1447 1447 1448 1448 1449 1449 1450 1450 1451 1451 Time [s] 27 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Speed [km/h] WPED [%] Power BSG [kw] Power [kw] Further work Powertrain energy flow complex torque-split situations 30.0 Electric drive initially 25.0 20.0 15.0 Lost energy for engine start can be determined 10.0 5.0 0.0-5.0-10.0 Evaluation of torque control during switch-over Power electric driv e (Front driv e-shaf ts + BSG) Ef f ectiv e Power ICE Power to Road (Wheels) Power to Wheels (Shaf ts) -15.0-20.0-25.0-30.0-35.0 Full description of complex deceleration condition with four negative torque components: ICE BSG Front electric machine 15.0 10.0 5.0 0.0-5.0-10.0-15.0 100 75 50 25 VSS WPED -40.0 Brake (mech. & regen.) 100 75 50 25 Mechanical brakes 0 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 0 Time [s] 28 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Speed [km/h] WPED [%] Power [kw] Power [kw] Further work Powertrain energy flow HV system efficiency estimation Efficiency of the HV system can be estimated by comparing the electromechanical power to the measured power at the HV battery Plot shows fixed h and auxiliary offset simplification, as h and auxiliary power are not constant Using an iterative approach, auxiliary power and h may be evaluated throughout the drivecycle 8 6 4 2 0 100 75 50 Power HV battery (from CAN) Power electric driv e (Front driv e-shaf ts + BSG) Power @HV battery Corrected HV Battery Power ( h = 73.5%, P auxiliary = 0.33kW) Corrected battery power to match electric propulsion value Power Losses & electric auxiliaries VSS WPED 26.0 24.0 22.0 20.0 18.0 16.0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0-2.0-4.0-6.0-8.0-10.0 100 75 50 25 25 0 0 1378.5 1379.0 1379.5 1380.0 1380.5 1381.0 1381.5 1382.0 1382.5 1383.0 1383.5 1384.0 1384.5 1385.0 1385.5 1386.0 1386.5 1387.0 1387.5 1388.0 1388.5 Time [sec] 29 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Speed [km/h] WPED [%] Power [kw] Power [kw] Further work Powertrain energy flow Correlation between wheel and shaft power 110 Excellent correlation between shaft and wheel torque measurement Wheel Power Shaf t Power 100 90 80 70 60 50 Deviation is in the region of 0.2 0.5 kw in normal driving conditions 40 30 20 10 0-10 As a result, the recuperation efficiency can be determined with a very high level of accuracy 3.0 2.0 1.0 0.0-1.0-2.0-3.0 Power dif f erence Wheel/Shaf t -20 Brake (mech. & regen.) 100.0 75.0 50.0 VSS WPED 100.0 75.0 50.0 25.0 25.0 0.0 0.0 3025 3030 3035 3040 3045 3050 3055 3060 3065 3070 3075 3080 3085 3090 3095 3100 3105 3110 3115 3120 3125 Time [s] 30 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

Summary The approach allowed a simplified analysis of energy flows to be established quite easily Charging and boosting energy flows could be determined Boosting energy can further be broken down into front axle and rear axle driving. Charging energy can be broken down into kinetic energy recovery, front axle charging, and ICE load point shifting Standardised tools and measurements for holistic powertrain analysis are essential for modern development processes 32 03/07/2018 Dr. Andrew Lewis 13 th International AVL Symposium on Propulsion Diagnostics

ANY QUESTIONS? Real-world to Lab Robust measurement requirements for future vehicle powertrains Andrew Lewis, Edward Chappell, Richard Burke, Sam Akehurst, Simon Pickering University of Bath Simon Regitz, David R Rogers Kistler Instrumente AG