Dynamic modeling and control of transcritical vapor compression system for battery electric vehicle thermal management

Garrow, Sarah Grace

Permalink

https://hdl.handle.net/2142/101473 Copy

Description

Title

Dynamic modeling and control of transcritical vapor compression system for battery electric vehicle thermal management

Author(s)

Garrow, Sarah Grace

Issue Date

2018-06-11

Director of Research (if dissertation) or Advisor (if thesis)

Alleyne, Andrew G.

Department of Study

Mechanical Sci & Engineering

Discipline

Mechanical Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• transcritical vapor compression system
• CO2 refrigerant
• battery electric vehicle
• thermal management
• model predictive control
• dynamic modeling

Abstract

Electrification is an increasing trend among vehicle systems such as aircrafts, heavy machinery, and civilian transportation. Battery electric vehicles (BEVs) are one such development that use a battery pack to generate electrical energy used to propel the vehicle and power its auxiliaries. However, the battery pack also generates thermal energy as a byproduct which affects the electrical performance of the battery pack. The inherent coupling between electrical and thermal performance creates a challenge in design and control of these complex systems. Furthermore, phase-out of common refrigerants drives interest in CO2 refrigerant, an environmentally friendly and safe alternative. However, these vapor compression systems operate transcritical, thus requiring novel control techniques. This thesis develops a framework for architecture and control design of BEV subsystems. The foundation of this process is the development of multi-domain models. Models for the transcritical vapor compression system and the vehicle cabin are derived from a first principles analysis. A model for a battery pack is derived from an equivalent circuit electrical model and a conservation of energy thermal model. All of the models capture dynamic, nonlinear behaviors important for control development and understanding of coupling between variables. Additionally, the models are scalable and able to be parameterized in order to represent many variations of system architectures. An air-cooled cabin and air-cooled battery pack configuration is demonstrated in open-loop and closed-loop simulations. For closed loop simulation, a model predictive controller (MPC) is compared to baseline decentralized, proportional-integral controllers. The model predictive control makes control decisions based on the minimization of a cost function that weights the regulation of specific variables (such as temperature of the battery pack and cabin) and power consumption of the actuators. It will be shown that the MPC, in the face of disturbances, is able to maintain outputs within their bounds while consuming less energy than baseline controllers.

Graduation Semester

2018-08

Type of Resource

text

Permalink

http://hdl.handle.net/2142/101473

Copyright and License Information

Copyright 2018 Sarah Garrow

Owning Collections