Welcome to the upgraded MacSphere! We're putting the finishing touches on it; if you notice anything amiss, email macsphere@mcmaster.ca

Dynamometer Characterization of Electric Powertrain Components for Accurate Modeling and Control Design of a Dual-Motor All-Wheel-Drive Electric Vehicle

dc.contributor.advisorEmadi, Ali
dc.contributor.authorAllca-Pekarovic, Alexander
dc.contributor.departmentMechanical Engineeringen_US
dc.date.accessioned2025-02-01T03:02:56Z
dc.date.available2025-02-01T03:02:56Z
dc.date.issued2025
dc.description.abstractUnited States government data shows AWD EV production increasing over 376 % from 2020 to 2023. The literature highlights energy savings from mixed-type dual EM powertrains and optimization-based energy management strategies (EMS), compared to single-type dual-motors and rule-based control. Other trends include the adoption of silicon carbide (SiC) based inverter devices and 800 V systems. Axial-flux machines have seen increased traction, with implementations in Ferrari’s SF90 Stradale, Lamborghini’s Revuelto, and Mercedes-Benz’s acquisition of YASA Ltd. These findings motivated the study of dual-motor AWD EV thermally constrained energy management. Using McMaster Automotive Resource Centre’s (MARC’s) facilities, accurate modeling of powertrain components was pursued to contribute realistic results. Firstly, inverter device materials and voltage ratings were studied in a Chevrolet Bolt EV model. Experimental validation was conducted on 1200 V inverters, powering 160+ kW traction machines. The model’s loss error was mostly within 100 W of measured loss. An empirical loss model revealed the analytical model estimates range within 6 km. This work highlighted the benefits of 800 V DC buses and SiC inverters. Secondly, experimental characterization of a yokeless and segmented armature (YASA) axial flux machine, by YASA Ltd., was documented. Dynamometer tests covered a wide torque, speed, and DC bus voltage range. The Bolt EV was modeled with the YASA machine, comparing its performance to the stock machine. All data was compiled and published in an online open-source repository. Lastly, thermally constrained energy management of various control strategies for a dual-motor AWD EV model were compared. Over two drive cycles, an MPC strategy ranked best in selected performance metrics. During a racetrack drive cycle, the MPC strategy kept the thermally limited motor cooler 246 % longer than a rule-based strategy. This work highlighted MPC’s potential in reducing total lifetime thermal wear of a dual-motor powertrain’s thermally limited motor.en_US
dc.description.degreeDoctor of Philosophy (PhD)en_US
dc.description.degreetypeThesisen_US
dc.description.layabstractThis thesis presents the experimental dynamometer work behind the accurate modeling of electric powertrain components, specifically the electric machine (EM) and inverter. The analytical inverter model is experimentally validated using electric machines as loads. The electric machine model is constructed from its experimental characterization data. Together, these models form a near-completely experimental-based electric drive unit (EDU). From this foundation, a dual-motor all-wheel-drive (AWD) electric vehicle (EV) model is built for the purpose of evaluating various control strategies’ thermally constrained energy management abilities. The control methods are ranked with respect to key performance indicators (KPIs) over the course of two drive cycles. Results from these drive cycles point to the model predictive control (MPC) strategy achieving the control objectives with the best quantified KPIs. Most importantly, it was able to keep the powertrain's thermally limited motor cooler 246 % longer than the second-best performing control strategy, a rule-based method applying a torque-split ratio algorithm.en_US
dc.identifier.urihttp://hdl.handle.net/11375/31014
dc.language.isoen_USen_US
dc.subjectDual-Motor AWD EVen_US
dc.subjectInverter Semiconductor Loss Modelingen_US
dc.subjectElectric Machine Characterizationen_US
dc.subjectThermally Constrained Energy Managementen_US
dc.subjectPowertrain Modelingen_US
dc.subjectDynamometer Testingen_US
dc.titleDynamometer Characterization of Electric Powertrain Components for Accurate Modeling and Control Design of a Dual-Motor All-Wheel-Drive Electric Vehicleen_US
dc.title.alternativePOWERTRAIN CHARACTERIZATION FOR MODELING AND CONTROL DESIGNen_US
dc.typeThesisen_US

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Allca-Pekarovic_Alexander_MJ_2025January_PhD.pdf
Size:
18.31 MB
Format:
Adobe Portable Document Format

License bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
1.68 KB
Format:
Item-specific license agreed upon to submission
Description: