Real-Time Energy Management and Transient Power Control for Fuel Cell Electrified Vehicles
Wu, Kai
2019
Abstract
Automotive OEMs have responded to energy and environmental concerns with mass-produced Hybrid Electric Vehicles (HEVs), Plug-in Hybrid Electric Vehicles (PHEVs), and Battery Electric Vehicles (BEVs) that satisfy various customers’ demands. While the sales volume of these vehicles continues to climb, OEMs recognize that Fuel Cell Vehicles (FCVs) could be the ultimate solution to electrification of personal transportation. Thus, they have forged ahead with developing commercial FCV technologies. However, several challenges exist in bringing Fuel Cell technology to mass production. Aside from steep costs, energy management for achieving total optimal system efficiency in real-time and under all driving conditions is still under development. There is room for improvement in controlling the transient power balance between the Fuel Cell System (FCS), high voltage battery, and driver demand, calling for a systematic framework and new tools to understand and address the FCS dynamic effects. This dissertation is devoted to providing a comprehensive framework for analyzing the dynamic effects of FCS on optimal energy management applications, and developing a hierarchical control framework for real-time energy management. Dynamic characteristics of a Proton Exchange Membrane Fuel Cell (PEMFC) system can impact fuel economy and load following performance of an FCV, especially if these dynamics are not considered when designing the top-level energy management strategy. To quantify the effects of FCS dynamics on optimal energy management, Dynamic Programming (DP) is adopted in this dissertation to derive optimal power split strategies at two levels: Level 1, where the FCS dynamics are ignored; and Level 2, where the FCS dynamics are incorporated. Analysis is performed to quantify the differences between these two strategies to understand the effects of FCS dynamics. The results show that ignoring slow FCS dynamics in DP can lead to several problems, including deteriorated power tracking, violation of charge sustaining performance, and loss of fuel economy. For the FCVs with fast power dynamics, an optimization-oriented supervisory controller based on Pontryagin's Minimum Principle (PMP) is proposed. The Adaptive-PMP (A-PMP) method inherits the advantages of model-based optimization to formulate a Hamiltonian and convert the trajectory optimization problem into pointwise-in-time optimization problem, where the co-state value is estimated and adapted based on average power and total travel time. A-PMP is evaluated on a high fidelity FCV powertrain model. Comparing to the default baseline energy management method, A-PMP yields better performance in fuel economy. Furthermore, a preliminary vehicle test shows up to $5.9%$ of improvement in fuel economy over an OEM's rule-based strategy. For the FCVs with slow power dynamics, an online energy management algorithm is proposed to mitigate the dynamic effects of FCS while maintaining a near-optimal fuel economy. The A-PMP-Model Predictive Control (APMP-MPC) scheme includes a top level power planning controller (A-PMP) and an intermediate level controller (MPC) to handle FCS transient dynamics. The proposed APMP-MPC is tested on an FCV powertrain model with simplified FCS dynamics. Simulation results demonstrate improvements in fuel economy on representative driving cycles under the assumption of prescient load information. Moreover, the effects of load prediction error on the APMP-MPC fuel economy performance are evaluated. Requirements on load prediction are identified to maintain the effectiveness of the proposed APMP-MPC algorithm for FCVs energy management. Finally, the sensitivity of the prediction error on fuel economy is shown to be attenuated by incorporating a rate limiter.Subjects
Real-time Energy Management Transient Power Control Fuel Cell Electrified Vehicles Hybrid Vehicles Model Predictive Control Optimization
Types
Thesis
Metadata
Show full item recordCollections
Remediation of Harmful Language
The University of Michigan Library aims to describe library materials in a way that respects the people and communities who create, use, and are represented in our collections. Report harmful or offensive language in catalog records, finding aids, or elsewhere in our collections anonymously through our metadata feedback form. More information at Remediation of Harmful Language.
Accessibility
If you are unable to use this file in its current format, please select the Contact Us link and we can modify it to make it more accessible to you.