東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Computationally Efficient Online Mod...

Liu, Ji.

Linked to FindBook

Google Book

Amazon

博客來

Computationally Efficient Online Model-Based Control And Estimation For Lithium-Ion Batteries.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Computationally Efficient Online Model-Based Control And Estimation For Lithium-Ion Batteries./
Author:	Liu, Ji.
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2017,
Description:	2017 p.
Notes:	Source: Dissertation Abstracts International, Volume: 78-11(E), Section: B.
Contained By:	Dissertation Abstracts International78-11B(E).
Subject:	Mechanical engineering. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10629081
ISBN:	9781369991840

Computationally Efficient Online Model-Based Control And Estimation For Lithium-Ion Batteries.
Liu, Ji.

Computationally Efficient Online Model-Based Control And Estimation For Lithium-Ion Batteries. - Ann Arbor : ProQuest Dissertations & Theses, 2017 - 2017 p.

Source: Dissertation Abstracts International, Volume: 78-11(E), Section: B.

Thesis (Ph.D.)--The Pennsylvania State University, 2017.

This dissertation presents a framework for computationally-efficient, health-conscious online state estimation and control in lithium-ion batteries. The framework builds on three main tools, namely, (i) battery model reformulation and (ii) pseudo-spectral optimization for (iii) differential flatness. All of these tools already exist in the literature. However, their application to electrochemical battery estimation and control, both separately and in an integrated manner, represents a significant addition to the literature. The dissertation shows that these tools, together, provide significant improvements in computational efficiency for both online moving horizon battery state estimation and online health-conscious model predictive battery control. These benefits are demonstrated both in simulation and using an experimental case study.

ISBN: 9781369991840Subjects--Topical Terms:

649730
Mechanical engineering.

Computationally Efficient Online Model-Based Control And Estimation For Lithium-Ion Batteries.
LDR:06363nmm a2200337 4500 001 2157891
005 20180608141653.5
008 190424s2017 ||||||||||||||||| ||eng d
020 $a 9781369991840
035 $a (MiAaPQ)AAI10629081
035 $a AAI10629081
040 $a MiAaPQ $c MiAaPQ
100 1 $a Liu, Ji. $3 3345707
245 1 0 $a Computationally Efficient Online Model-Based Control And Estimation For Lithium-Ion Batteries.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2017
300 $a 2017 p.
500 $a Source: Dissertation Abstracts International, Volume: 78-11(E), Section: B.
500 $a Adviser: Hosam K. Fathy.
502 $a Thesis (Ph.D.)--The Pennsylvania State University, 2017.
520 $a This dissertation presents a framework for computationally-efficient, health-conscious online state estimation and control in lithium-ion batteries. The framework builds on three main tools, namely, (i) battery model reformulation and (ii) pseudo-spectral optimization for (iii) differential flatness. All of these tools already exist in the literature. However, their application to electrochemical battery estimation and control, both separately and in an integrated manner, represents a significant addition to the literature. The dissertation shows that these tools, together, provide significant improvements in computational efficiency for both online moving horizon battery state estimation and online health-conscious model predictive battery control. These benefits are demonstrated both in simulation and using an experimental case study.
520 $a Two key facts motivate this dissertation. First, lithium-ion batteries are widely used for different applications due to their low self-discharge rates, lack of memory effects, and high power/energy densities compared to traditional lead-acid and nickel-metal hydride batteries. Second, lithium-ion batteries are also vulnerable to aging and degradation mechanisms, such as lithium plating, some of which can lead to safety issues. Conventional battery management systems (BMS) typically use model-free control strategies and therefore do not explicitly optimize the performance, life span, and cost of lithium-ion battery packs. They typically avoid internal damage by constraining externally-measured variables, such as battery voltage, current, and temperature. When pushed to charge a battery quickly without inducing excessive damage, these systems often follow simple and potentially sub-optimal charge/discharge trajectories, e.g., the constant-current/constant-voltage (CCCV) charging strategy. While the CCCV charging strategy is simple to implement, it suffers from its poor ability to explicitly control the internal variables causing battery aging, such as side reaction overpotentials. Another disadvantage is the inability of this strategy to adapt to changes in battery dynamics caused by aging.
520 $a Model-based control has the potential to alleviate many of the above limitations of classical battery management systems. A model-based control system can estimate the internal state of a lithium-ion battery and use the estimated state to adjust battery charging/discharging in a manner that avoids damaging side reactions. By doing so, model-based control can (i) prolong battery life, (ii) improve battery safety, (iii) increase battery energy storage capacity, (iv) decrease internal damage/degradation, and (v) adapt to changes in battery dynamics resulting from aging. These potential benefits are well-documented in the literature. However, one major challenge remains, namely, the computational complexity associated with online model-based battery state estimation and control. The goal of this dissertation is to address this challenge by making five contributions to the literature. Specifically: • Chapter 2 exploits the differential flatness of solid-phase lithium-ion battery diffusion dynamics, together with pseudo-spectral optimization and diffusion model reformulation, to decrease the computational load associated with health-conscious battery trajectory optimization significantly. This contribution forms a foundation for much of the subsequent work in this dissertation, but is limited to isothernal single-particle battery models with significant time scale separation between anode- and cathode-side solid-phase diffusion dynamics. • Chapter 3 extends the results of Chapter 2 in two ways. First , it exploits the law of conservation of charge to enable flatness-based, health-conscious battery trajectory optimization for single particle battery models even in the absence of time scale separation between the negative and positive electrodes. Second, it performs this optimization for a combined thermo-electrochemical battery model, thereby relaxing the above assumption of isothermal battery behavior and highlighting the benefits of flatness-based optimization for a nonlinear battery model. • Chapter 4 presents a framework for flatness-based pseudo-spectral combined state and parameter estimation in lumped-parameter nonlinear systems. This framework enables computationally-efficient total least squares (TLS) estimation for lumped-parameter nonlinear systems. This is quite relevant to practical lithium-ion battery systems, where both battery input and output measurements can be quite noisy. • Chapter 5 utilizes the above flatness-based TLS estimation algorithm for moving horizon state estimation using a coupled thermo-electrochemical equivalent circuit model of lithium-ion battery dynamics. • Chapter 6 extends the battery estimation framework from Chapter 5 to enable moving horizon, flatness-based TLS state estimation in thermo-electrochemical single-particle lithium-ion battery models, and demonstrates this framework using laboratory experiments.
520 $a The overall outcome of this dissertation is an integrated set of tools, all of them exploiting model reformulation, differential flatness, and pseudo-spectral methods, for computationally efficient online state estimation and health-conscious control in lithium-ion batteries.
590 $a School code: 0176.
650 4 $a Mechanical engineering. $3 649730
650 4 $a Systems science. $3 3168411
650 4 $a Artificial intelligence. $3 516317
690 $a 0548
690 $a 0790
690 $a 0800
710 2 $a The Pennsylvania State University. $b Mechanical Engineering. $3 2103981
773 0 $t Dissertation Abstracts International $g 78-11B(E).
790 $a 0176
791 $a Ph.D.
792 $a 2017
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10629081