«A Dissertation Presented to The Academic Faculty by Thomas Heenan Bradley In Partial Fulfillment of the Requirements for the Degree Doctor of ...»
MODELING, DESIGN AND ENERGY MANAGEMENT OF FUEL
CELL SYSTEMS FOR AIRCRAFT
The Academic Faculty
Thomas Heenan Bradley
In Partial Fulfillment
of the Requirements for the Degree
Doctor of Philosophy in the
School of Mechanical Engineering
Georgia Institute of Technology
Copyright © Thomas Heenan Bradley 2008
MODELING, DESIGN AND ENERGY MANAGEMENT OF FUEL
CELL SYSTEMS FOR AIRCRAFT
Dr. David E. Parekh, Advisor Dr. William J. Wepfer School of Mechanical Engineering School of Mechanical Engineering Georgia Institute of Technology Georgia Institute of Technology Dr. Dimitri N. Mavris Dr. Yogendra Joshi School of Aerospace Engineering School of Mechanical Engineering Georgia Institute of Technology Georgia Institute of Technology Dr. Thomas F. Fuller School of Chemical and Biomolecular Engineering Georgia Institute of Technology Date Approved: August 01, 2008
ACKNOWLEDGEMENTSThis research was funded in part by the NASA University Research Engineering Technology Institute (URETI) grant to the Georgia Institute of Technology.
Thanks to my academic advisors David Parekh, Tom Fuller and Dimitri Mavris. I am grateful for all of your generosity in terms of resources and attention and for your dedication to a creative and collaborative learning environment. It was pleasure to be to able to engage with creative and knowledgeable people who work with such infectious enthusiasm. Thanks also to committee members Bill Wepfer and Yogendra Joshi.
Thanks to Blake Moffitt for his tireless collaboration and for providing my introduction to the world of aeronautical engineering. This work would have been impossible without his many contributions.
Thanks to the ASDL for making me one of their own. Thanks to the members of Dr. Fuller’s lab for friendship and the inspiration that comes with the most interesting and cross-disciplinary lecture series on campus. Thanks to the GTRI Fuel Cell and Battery Technology faculty, Comas Haynes and Gary Gray for thousands of instances of teaching and research help.
Thanks to Kimberly, without whom none of this could have happened.
Atlanta, Georgia July 25, 2008 iv
TABLE OF CONTENTSPage
TABLE OF CONTENTS vLIST OF TABLES
3.2 Research Question 1 – Fuel Cell Systems Modeling and Validation 32
3.3 Research Question 2 – Fuel Cell Aircraft Integrated Design Studies 34
3.4 Research Question 3 – Energy Management Studies for Fuel Cell Hybrid
5.1.2. Results and Discussion for Fuel Cell System Design Rules Comparison 84
5.2 Application-level and Powerplant-level Design Metric Comparisons 88
6.3.2. Energy Management for Level Flight with Random Disturbance 120 6.3.3. Hybridization for Cyclical Power Missions and Level Flight 121 6.3.4. Hybridization for Missions with a High Power Climb Followed by
Table 3. Characteristics of conceptual compressed hydrogen storage system 54 Table 4.
Primary design variables and side constraints for conceptual design of a fuel cell
Table 8. Comparison of aircraft characteristics for Aircraft A and Aircraft B 85 Table 9.
Tabular summary of steps associated with the endurance design metric
Table 18. Values and uncertainty for the primary data acquired during testing 174 Table 19.
Comparison of electrochemical powerplants for long range and long endurance
Table 21. Aircraft performance sensitivities to power and propulsion system performance 195 Table 22.
Wiring spreadsheet and sensor list for fuel cell A/D converter and system
Figure 6. Scale comparison of small-scale fuel cell powered UAVs constructed to date 20 Figure 7.
Visual representation of design methods comparison associated with
Figure 9. Summary of research questions and tasks associated with this dissertation 42 Figure 10.
Summary of development tasks associated with this dissertation 43
Figure 18. Canonical multidisciplinary design and optimization problem structure 57 Figure 19.
Default design structure matrix for fuel cell UAV design problem 58 Figure 20. Multi-objective tradeoff study of fuel cell aircraft performance as a function of
Figure 25. Actual by predicted plots for the Hydrogen Utilization contributing analysis 69 Figure 26.
Actual by predicted plots for the balance of plant contributing analysis 70 Figure 27. Actual by predicted plots for the Hydrogen Tank Mass and Dimensions
Figure 36. Comparison between rules based and integrated design of fuel cell powerplant 87 Figure 37.
Efficiency-based comparison between rules based and integrated design of
Figure 57. Optimal energy management strategy for hybrid fuel cell powered aircraft during level flight with burst power demands and a charge sustaining strategy 123 Figure 58.
Optimal energy management strategy for hybrid fuel cell powered aircraft during level flight with burst power demands and a charge depleting strategy 124 Figure 59. Optimal energy management strategy for hybrid fuel cell powered aircraft during level flight with burst power demands and a charge sustaining strategy 125 Figure 60. Optimal periodic flight paths for fuel cell and internal combustion powered
Figure 63. Fuel cell powered aircraft constructed for validation of design methodology 132
Figure 73. Representative flight test results for fuel cell powered circuit flight 150 Figure 74.
Representative flight test results for fuel cell powered straight-line flight 151
Figure 84. Hardware simulation performance during takeoff and climb flight segments 175 Figure 85.
Hardware simulation performance during the entire long endurance flight 176
Figure 90. System diagram showing power and signal communication between modules 198
Figure 94. Dynamic performance of the compressor control system during flight testing 206 Figure 95.
Dynamic performance of the compressor control system during benchtop
Figure 99. Radiator configurations tested for use in the FC UAV demonstration aircraft (a) Brass Radiator, (b) Carbon Foam I radiator with 0.
125 in. pin hole fins, (c) Carbon
Figure 105. Sample data acquisition results from Horizon H300 with GTRI controller 218 Figure 106. Comparison of fuel cell polarization curves with and without control
C = sequential unconstrained minimization technique tuning variable CD = aircraft coefficient of drag CL = aircraft coefficient of lift Cq = propeller coefficient of torque Crr = coefficient of rolling resistance CT = propeller coefficient of thrust
mH2tank = mass of the hydrogen storage tank, kg mliner = hydrogen tank liner mass, kg mpower = powerplant mass including powertrain, fuel and tankage, kg
PH2 = hydrogen pressure in fuel cell anode manifold, Pa pm = cathod manifold pressure, Pa Q = propeller and electric motor torque, Nm qH2 = flow rate of hydrogen, kg/sec qH2 = cathode resistance limited flow rate of hydrogen, kg/sec R = optimization objective function
V° = Nernst potential of the hydrogen oxidation reaction, 1.23V vdesired = desired aircraft velocity, m/s verror = error in aircraft velocity, m/s vpurge = volume of hydrogen purged in a purge cycle, L vconsumed = volume of hydrogen consumded in a purge cycle, L
Fuel cell powered aircraft have been of long term interest to the aviation community because of their potential for improved performance and environmental compatibility.
Only recently have improvements in the technological readiness of fuel cell powerplants enabled the first aviation applications of fuel cell technology. Based on the results of conceptual design studies and a few technology demonstration projects, there has emerged a widespread understanding of the importance of fuel cell powerplants for nearterm and future aviation applications. Despite this, many aspects of the performance, design and construction of robust and optimized fuel cell powered aircraft have not been fully explored.
This goal of this research then is to develop an improved understanding of the performance, design characteristics, design tradeoffs and viability of fuel cell powerplants for aviation applications. To accomplish these goals, new modeling, design, and experimental tools are developed, validated and applied to the design of fuel cell powered unmanned aerial vehicles.
First, a general sub-system model of fuel cell powerplant performance, mass and geometry is derived from experimental and theoretical investigations of a fuel cell powerplant that is developed in hardware. These validated fuel cell subsystem models are then incorporated into a computer-based, application-integrated, parametric, and optimizeable design environment that allows for the concurrent design of the aircraft and fuel cell powerplant. The advanced modeling and design techniques required for modern aircraft design (including multi-disciplinary analysis, performance optimization under
level and are linked to aircraft performance and design metrics. These tools and methods are then applied to the analysis and design of fuel cell powered aircraft in a series of case studies and design experiments.
Based on the results of the integrated fuel cell system and aircraft analyses, we gain a new understanding of the interaction between powerplant and application for fuel cell aircraft. Specifically, the system-level design criteria of fuel cell powerplants for aircraft can be derived. Optimal sub-system configurations of the fuel cell powerplant specific to the aircraft application are determined. Finally, optimal energy management strategies and flight paths for fuel cell and battery hybridized fuel cell aircraft are derived.
The results of a series of design studies are validated using hardware in the loop testing of fuel cell propulsion systems and field testing of a series of fuel cell powered demonstrator aircraft.
The focus of this dissertation is the modeling, design and energy management of fuel cell powerplants for aircraft. This chapter presents an introduction and motivational background to the topics of systems modeling and design, and fuel cell powerplants.
1.1 Systems Modeling and Design The modeling and design aspects of this investigation build on the tools of multidisciplinary analysis and design as exercised in the aerospace design community.
Any non-trivial design process consists of numerous processes that exhibit varying degrees of interconnection and interrelation [1,2]. Traditionally, this design process had to be handled by a single expert designer who had enough experience in the entire problem domain so that the designer was able to guide the design through decision making. As the complexity of design has increased with increasing scale, increasing scope, incorporation of uncertainty, design for constraint robustness, and multi-stage decision making processes, the requirements of a designer have increased so that no one person can perform satisfactorily.
Multidisciplinary analysis and design have evolved to enable the analysis and design of complex systems. The tools of multidisciplinary design allow for the decomposition of a monolithic and integrated design/analysis problem into a series of independent sub-processes with defined inputs, outputs and interconnections between the sub-processes. The casting of a design problem into this multi-disciplinary analysis form is generally a subjective task, that must informed by knowledge of the information that is
support, of design space exploration must all have difference decomposition form. The determination and defense of the form of the multi-disciplinary analysis and design problem is a central problem in systems modeling and design.
Optimal design problems consist of choosing the design parameters of the multidisciplinary analysis so as to maximize a design objective subject to constraints .
The design then goes from conceptual design, where the design exists entirely in models, to detail design, where the physical, realizeable specifications of the components have been made . The process of going from conceptual to detail design for complex and multidisciplinary systems is a developing field of system design.