Open Collections

Ferreira da Silva, Gabriel

University of British Columbia

Transportation electrification is at the core of the possible solutions to many challenges the world is currently facing. Efficient vehicle electrification has the potential to simultaneously reduce greenhouse gasses emissions and to tackle range anxiety issues. Among different strategies for advancements in Electric Vehicle (EV) efficiency, enhancing Regenerative Braking (REGEN) capabilities is an area with opportunities. As REGEN faces different impediments, upgrades in safety, efficiency, and/or battery quality of life are usually accompanied with further strain in energy management schemes, limiting REGEN performance. Power Electronics (PE) improvements are among the options that have the potential to benefit REGEN and overall efficiency. This work proposes a method to improves REGEN without adding extra stress on the other aspects that limit its performance, by optimizing PE-stage switch selection using openly available, manufacturer-provided data. To do so, the thesis develops a flexible simulation platform capable of: 1) integrating various subsystem modeling approaches, 2) analyzing different EV configurations, architectures, and components, and 3) analyzing the dynamic behavior of the Battery Electric Vehicle (BEV) while maintaining low simulation time. It also adopts a multiobjective optimization approach that gives the user freedom to define the weight of the objectives, as well as to include new objectives at any time - as long as the initial design choices do not change. The combination of a simulation platform suited for model-based design and an optimization formulation yields a method that fits well within the Design Automation (DA) framework. Therefore, the thesis is constructed with the framework as a guideline. The simulations show that proper switch selection can improve REGEN by over 18% and EV range efficiency by over 20%. The solution is corroborated by the results of the sensitivity and the robustness analysis.

Text

2021-07-30

Attribution-NonCommercial-ShareAlike 4.0 International

http://hdl.handle.net/2429/79169

Electrical and Computer Engineering

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-sa/4.0/