You are on page 1of 18

3rd International Energy Conversion Engineering Conference 15 - 18 August 2005, San Francisco, California

AIAA 2005-5574

A Generalized Aircraft Sizing Method and Application to Electric Aircraft


Taewoo Nam , Danielle S. Soban , and Dimitri N. Mavris
Georgia Institute of Technology, Atlanta, GA 30332-0150
Internal combustion (IC) engines, which consume hydrocarbon fuels, have dominated the propulsion systems of air-vehicles during a century of aviation history. In the past decade, however, a combination of environmental, technological, and socio-economic changes have stimulated the search for new, alternative sources of power that could challenge the dominance of the IC engine. In particular, fuel cells are increasingly being considered as an alternate power source due to their potential advantages over the traditional power system. Nevertheless, traditional aircraft sizing methods currently employed in the conceptual design phase are not immediately applicable to such revolutionary power drive aircraft designs. Motivated by such deciencies in state-of-the art sizing methods, a generalized aircraft sizing method has been developed as a solution to this challenge. A brief outline of the method and preliminary results from its application to an electric high altitude long endurance (HALE) conguration are provided in this paper.

Nomenclature
CD CDo CDR D E ECE ENE go h K1 K2 m n nT nCE nNE nPD P p po poSL q R s SG
Graduate Research

Coecient of clean conguration drag Zero lift drag coecient Coecient of additional drag Basic conguration drag Amount of stored onboard energy Amount of consumable onboard energy Amount of non-consumable onboard energy Gravity constant Altitude Drag polar coecient for 2nd order term Drag polar coecient for 1st order term Number of total mission segments Load factor Number of total power paths Number of consumable power paths Number of non-consumable power paths Number of power devices Power available Power available of individual power path Prototype power of individual power path Prototype power of individual power path at sea level static Dynamic pressure Additional drag to D Travel distance Ground roll

Research Assistant, School of Aerospace Engineering, and AIAA Student Member. Engineer, School of Aerospace Engineering, and AIAA Member. Boeing Professor, Director ASDL, School of Aerospace Engineering, and AIAA Senior Member.

1 of 18 American Institute of Aeronautics and Astronautics

Copyright 2005 by Taewoo Nam, Danielle S. Soban and Dimitri N. Mavris. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission.

T u V Vstall VTO W WCE Wenergy WE WNE WP WPD WTO ze


CE NE

CE NE PD CE NE i CE

Thrust Drag to thrust ratio Free stream velocity Stall speed Take-o speed Instantaneous aircraft weight Consumable energy weight Total stored energy weight Empty weight Non-consumable energy weight Payload weight Propulsion system weight Take-o gross weight Energy height Power lapse ratio Weight fraction Weight correction factor Consumable energy allowance ratio Non-consumable energy allowance ratio Empty weight fraction Ratio of empty weight less the weight of installed propulsion systems to the take-o gross weight Stored product to fuel ratio Specic energy of consumable energy Specic energy of non-consumable energy Specic power of power devices Consumable energy weight fraction Non-consumable energy weight fraction Power devices weight fraction Overall eciency of individual power path Free-stream air density Power fraction of individual power path Weight-specic mechanical energy of aircraft Overall power specic fuel consumption

Superscript i Number of a power path powered by consumable energy sources j Number of a power path powered by non-consumable energy sources (s) Mission segment number

I.

Introduction

the Wright brothers rst ight, been by internal engines Since consume always apparent factOnlymost aircraft havepoweredpoweredthe aerospacecombustiondierent that hydrocarbon fuels. a few attempts have reminded community of the obvious but not that aircraft can be by dierent energy sources or power generation devices. One alternative energy source envisioned for aircraft propulsion was atomic energy. Development of a nuclear-turbojet powered airplane was initiated by the Nuclear Energy for the Propulsion of Aircraft (NEPA) Project, launched by the United States Air Force (USAF) in 1946, and the follow-up Aircraft Nuclear Propulsion (ANP) Program, controlled by the joint Atomic Energy Commission (AEC) and USAF.1 By the time President Kennedy delivered a statement canceling the program on March 28, 1961, about $1 billion had been devoted to the development of a nuclear-powered aircraft over nearly 15 years.2 However, no airplane powered by a nuclear propulsion system ever ew. Solar-powered aircraft have also attracted the attention of several agencies over the past decades because of their promising potential in military and civilian applications. The most noteworthy examples are Dr. Paul MacCreadys Solar Challenger, which crossed the English Channel on July 7, 1981, on solar power alone. The research community also introduced an interesting nonchemical propulsion concept of beaming power, which
2 of 18 American Institute of Aeronautics and Astronautics

is based on wireless transmission of electrical energy, and thus ends up with a nearly fuelless and emissionfree ight with extremely long endurance capability. Motivated by such potential advantages, researchers in several countries ew a variety of model aircraft using beamed microwave energy twenty years ago.3 However, the nature of the microwave beam causes it to dissipate with distance, resulting in a commensurate decrease in power delivered to the target. With only lasers left as a potentially feasible option for practical power beaming, a group of researchers proposed a concept of a laser-powered transportation system.4 Despite all such earlier eorts to develop alternate energy sources, none had paved the way to development of any practical, unconventionally-powered aircraft, and the dominance of IC engines was deemed unbeatable. However, capitalizing on enormous advances in energy storage and conversion and other improvements during the past decades, alternate energy sources and revolutionary propulsion systems are attracting renewed attention. Nuclear power has been envisioned in power plants for extremely long endurance reconnaissance ights5 as well as space exploration.6, 7 Beamed power was also rigorously reinvestigated for many potential aerospace applications, including high-altitude airships, extra-terrestrial robotic rovers and aircraft, and small or swarming unmanned aircraft.3, 6 Solar energy has been continuously investigated, envisioned for high altitude and long endurance applications such as the AeroVironment Helios, which has recently established a new altitude record of 96,500 feet and which has been deemed as a possible alternative to communication satellites. Fuel cells, which have already been used to power electrical systems of spacecraft on every manned space ight by NASA, began to be considered for aeronautical application due to signicant improvements in specic power and power density in recent years.8 This rigorous search for alternative energy sources and revolutionary propulsion systems has been stimulated by increased public concern about the environmental impact of engine emissions. From 1998 to 1999, a series of high-level studies exploring aircraft with unconventional propulsion systems that focused on reducing aircraft emissions and noise was jointly performed by NASA Langley Research Center (LaRC) and NASA Glenn Research Center (GRC).9 Such eorts culminated in the Quiet Green Transport (QGT) study, one of the initial studies undertaken in the Revolutionary Aerospace Systems Concepts (RASC) Program.9, 10 Building upon this momentum, NASA has recently identied the low emissions alternative power (LEAP) project, which is ultimately pursuing zero emissions aircraft as one of the four exclusive projects under the Vehicle Systems Program (VSP).11 In addition, development of revolutionary aerospace concepts that are not feasible with conventional propulsion systems such as Helios and the vehicles designed for planetary exploration, has provided an outlet for revolutionary propulsion technologies. Furthermore, a continuous decrease in alternative energy costs due to technological advances in concert with a steep increase in oil prices in recent years has been pushing the advent of economically viable alternative energy sources forward. Finally, glowing concerns about national energy security have driven the governments of many countries to reduce their dependence on imported energy sources and to support cutting edge research to develop sustainable energy sources. As a response to these trends, a considerable volume of research dealing with the design and analysis of alternative propulsion systems has been published. Nevertheless, it is the authors observation that most system-level research has simply applied revolutionary propulsion system technologies to conventional architectures while less research has been devoted to exploring revolutionary aircraft concepts with revolutionary propulsion systems. One of possible reasons for such limitations is that the traditional aircraft sizing method is not immediately applicable to revolutionary concepts that involve unconventional energy sources. Most recent system-level studies1215 evaluate unconventional propulsion systems by retrotting an existing aircraft and assessing the impact on aircraft performance without sizing the aircraft. The QGT study10, 16 presented unconventionally powered aircraft sizing practices for two specic congurations: a conventional wing-tail combination with over-wing, hydrogen-fueled engines and a blended-wing-body concept powered by hydrogen fuel cells. The analyses were achieved with signicant modications of the traditional sizing code, the Flight Optimization System (FLOPS) code. Aircraft sizing is a critical aspect of system-level study because the aircraft sizing process is a prerequisite task of most design and analysis activities, including internal layout, cost analysis, and system eectiveness analysis. For instance, one of the results of aircraft sizing, the initial estimation of thrust or power required, is a primary input to the preliminary investigation of the engine company, particularly if a new propulsion system is jointly developed. However, a comprehensive, structured, and generalized aircraft sizing method that is applicable to a wide range of unconventionally powered aircraft has not yet been fully developed. Motivated by such deciencies, the authors proposed a power-based aircraft sizing formulation applicable to unconventional energy-consuming concepts at the 44th AIAA Aerospace Sciences Meeting and Exhibit.17

3 of 18 American Institute of Aeronautics and Astronautics

Notional Concept

Aerodynamics Propulsion Systems Empirical Weight Equation

Point Performance Requirements

Constraints Analysis

Mission Analysis

Mission Performance Requirements


Mission Range Endurance Payload etc.

TOFL Climb Acceleration Cruise Sustain Turn Service Ceiling etc.

Thrust to Weight Ratio Wing Loading

Fuel Fraction

Weight Estimation

Weight

Sized Configuration
Wing Area Thrust Weight

Figure 1. Traditional aircraft sizing process.

In a follow-up eort, this article presents the renement of the formulation and its application to a HALE conguration as a proof of the concept.

II.

Traditional Sizing Methods

Before the deciencies of the traditional method and the need for a new method can be identied, aircraft sizing should be dened and and reviewed. According to several aircraft design textbooks1820 that presents a detailed process associated with tradition aircraft sizing methods, the sizing process is initiated with a notional air-vehicle concept, as depicted in Figure 1. The goal of aircraft sizing is not to yield an optimum aircraft design, but rather, according to the literal meaning of the word size, to determine two scales of the given concept: a geometric scale that is dictated by the wing area and a propulsive scale that is dictated by the amount of thrust of the engine. Traditionally, the scales are determined by establishing two balances: required thrust versus available thrust and required fuel versus available fuel. Thrust balance is achieved by a process known as the constraint analysis. Most aerodynamic point performance requirements, which include take-o eld length, climb, acceleration, sustained turn, and approach speed, can be expressed as functions of thrust loading (T/W) at sea level and wing loading (W/S) for any given aircraft geometry, thrust lapse behavior, and ight conditions. Thus, the feasible solution region can be identied by a set of constraint curves that represent each performance requirement. In general, the best values for T/W and W/S can be obtained from the location of the lowest thrust loading, since any redundant thrust will most likely increase the propulsion system weight and thus, the take-o gross weight. Fuel balance is achieved by a process known as the mission analysis, which determines the required fuel fraction by calculating the product of all the weight fractions of

4 of 18 American Institute of Aeronautics and Astronautics

discretized mission segments. The beauty of using non-dimensional parameters that are normalized by the take-o gross weight is that both power and fuel balance can be established by an independent analysis of the aircraft weight. Thus, the constraint analysis and the mission analysis do not need to be repeated, and the iteration process can be restricted to the inside of the weight estimation module. This benet may not be noticeable after an one-time run, but it would be substantial for such an application as an optimization problem or a stochastic simulation. The results from the two analyses are fed into the weight estimation module. Combining this information with the weight regression equations, the take-o gross weight is obtained. Once the take-o gross weight as well as the T/W and W/S values are available, the amount of thrust and the wing area of the notional conguration can be established.

III.

Fundamentals of the Proposed Formulation

The overall process of the traditional aircraft sizing method briey outlined above can be applied to revolutionary aerospace vehicle concepts that call for the consumption of unconventional types of energy. In order to do so, a considerable modication of the formulation is required. This section discusses the shortfalls of the traditional methods and four key ideas implemented in the proposed method. A. Generalized Propulsion System

A propulsion system is a device that produces propulsive thrust through a series of energy conversions. This process is typically aected by the design parameters of the system and operating conditions such as Mach number, altitude, and ambient temperature. Because the technology behind the traditional air-breathing combustion engines has matured, thrust lapses and SFC behaviors of traditional air-breathing combustion engines are well understood. Thus, aircraft design engineers need not track every detail of the internal energy conversion process inside the engine. Instead, they can obtain the thrust lapse, SFC behavior, and the scaling laws of a notional engine from well-established historical data by selecting the most suitable engine type and cycle, which depend on the mission requirements. In addition, the propulsion system data can also be created from physic-based legacy codes in which substantial knowledge obtained from several decades of engine development experiences was implemented. In contrast, emerging revolutionary propulsion systems have not yet been fully developed. They are continuously evolving, and thus, their thrust and fuel consumption behavior and scaling laws are not well established. Furthermore, several emerging energy sources introduce more ambiguous boundaries between the airframe design and the propulsion system design. For example, the selection and sizing process of a fuel cell system is a signicant multidisciplinary challenge in itself. Therefore, a more generalized technique for modeling the propulsion system is desired. Generalized modeling starts with decomposition of the system. A propulsion system consists of a series of power generation or conversion devices, as depicted in Figure 2. Such devices can be categorized into the power generation device (PGD), the power transformation device (PTD), and nally, the power output device (POD), the last power transformation device. Then, the most interesting parameters of propulsion systems in view of aircraft sizing, output power, fuel consumption, and system weight, are dictated by the following component level parameters: specic energy of the energy sources, specic power of each power device, and the eciency of each power device. Specic energy, E , is the contained energy per unit weight. The specic power of a power device, DV , is the amount of output power produced by the device per unit weight. The eciency, , of the device is the ratio of the amount of output energy to input energy of each
Pext

Energy (Fuel)
Power
PO

PGD
po

PTD1
1
p1

...
1 0 PO
0 k 1

PTDk

pk
0 k

...

PTDn-1
n-1
p n 1

POD
n
pn

0 PO

( ) PO

Figure 2. Generalized propulsion system model.

5 of 18 American Institute of Aeronautics and Astronautics

( ) PO

0 n 2

( ) PO

0 n 1

( ) PO

0 n

( ) PO

device. Then, the nal propulsive power, P , is given as P = n n1 1 0 Po = ()Po (1)

where is dened as a product operator that multiplies all values associated with the variable, and Po is the rst power input from the energy source. Fuel consumption, WF , is given as WF = Finally, the power system weight is given as
nPD nPD

P E ()

(2)

WPD =
k=1

WPDk =
k=1

P PDk
nk ()

(3)

Therefore, if the specic energy of the energy source and the specic power and the eciency of each power device are established, then the system characteristics and behavior can be fully described for the purpose of aircraft sizing. Most aircraft propulsion systems provide the power required to other subsystems such as the cooling system, the hydraulic system, and the electric system as well as propulsive power by means of engine bleed air and mechanical power extraction. In order to capture the loss due to power extraction, a small modication to the previous equation is needed. If Pext is taken out at the k th power device, then the modied eciency, k , of the power device can be expressed as follows: k = (k ) (4)

where is the ratio of the amount of power extraction to the amount of input power of the power device. B. Consideration of Multiple Power Paths

This formulation needs be further generalized in order to handle such cases in which either multiple propulsion systems are installed in aircraft or multiple energy sources are used. Except for a few aircraft such as KB-50J, a six-engine aerial refueling tanker equipped with four prop engines and two turbojet engines,21 most existing aircraft are equipped with a single engine or multiple identical engines. Therefore, the thrust available and fuel consumption for most conventional aircraft can be established by one engine deck. However, if dierent types of propulsion systems and energy sources are equipped, and the power contribution rate of each propulsion system varies with ight conditions, then the traditional sizing formulation cannot handle this situation properly. A mathematical representation for multiple power or energy sources can be developed by introducing the concept of power path. A power path is dened as a set of power devices along which a series of energy conversion processes take place. The vehicle is powered by nT dierent power paths, each consisting of multiple energy sources and energy transformation devices (ETD). Then, the power available to the aircraft is given as the sum of the power from all individual power paths.
nT

P =
i=1

(5)
i

This expression can be modied by introducing a power fraction factor, p where,


nT i

, as follows: (6)

i=1

= 1, and
i

>0

(7)

The rst power input from the original energy source, po , is eventually converted to the nal propulsive power through multiple energy transformation processes. The nal output power is given as the product of i the eciencies and po to account for losses associated with each transformation/conversion. p
i

= i poi
6 of 18

(8)

American Institute of Aeronautics and Astronautics

In order to simplify the formulation, it would be better to establish a reference power for each power path. The reference power is determined by the maximum rst power input from the energy source at sea level, known as poSL , by introducing a power lapse ratio, . poi = poi SL Finally, combining Eq. (6), Eq. (8), and Eq. (9) yields poi = SL C.
i i i

(9)

P
i

(10)

Consideration of Multiple Energy Sources

In order to y the mission, an aircraft must store sucient energy onboard. Conventional stored energy is in the form of hydrocarbon fuel, whose weight is consumable during ight. However, some emerging energy sources such as an electric battery and perhaps nuclear power maintain virtually constant weight during energy conversion processes. In this study, consumable energy is dened as energy that is derived from a source whose weight is reduced during power generation such as traditional hydrocarbon fuels. Alternatively non-consumable energy is dened here as energy derived from a source whose weight stays constant or changes negligibly during power generation such as an electric battery, human power, or a nuclear battery. In order to produce a generalized formulation, the aircraft is assumed to have multiple power paths that have either consumable or non-consumable energy sources. Then, the total energy stored onboard, E, is the sum of these two types of energy as follows:
nCE

E=
i=1

ECE +
j=1

nNE

ENE

(11)

where nCE is the number of power paths of consumable energy, and nNE is the number of power paths of non-consumable energy. Therefore, the total onboard stored energy weight, Wenergy is
nCE

Wenergy = WCE + WWE =


i=1

i WCE

nNE

+
j=1

WNE

(12)

It must be noted that this formulation calculates the energy weights in terms of power paths and not energy types. For instance, if JP-8 is used for both a conventional jet engine and a fuel cell system that powers electric motors and propellers, then this system has two power paths, and the required fuel weight for each of the two power paths is estimated separately. As new energy sources start to be utiFavorable to Range and Endurance lized, hybrid propulsion systems and/or a combination of dierent types of propulsion systems that utilize multiple energy sources, may power future aircraft. Each of the power devices and their associated energy sources have dierent characteris- Figure 3. Comparison of specic energy and specic power for tics in terms of specic power and specic various power source technologies22 energy. To reduce the weight of an aircraft, higher specic power and specic energy are always desired, but nature does not allow both in one source, as depicted in Figure 3. In general, higher specic energy may be favorable for range and endurance capability and higher specic power may be favorable for maneuvering capability. When multiple choices are available, designers may want to optimize the propulsion system architecture by properly mixing dierent
7 of 18 American Institute of Aeronautics and Astronautics
Favorable to Maneuvers

types of power devices and energy sources. For instance, in the case of sizing an aircraft that is required to perform high-power demanding maneuvers for short periods during a mission, then adding a booster power system that has higher specic power yet lower specic energy would be a better approach than sizing the entire propulsion system with one combination of a power device and an energy source at the critical condition, as demonstrated in the previous research.5, 14 However, the decision must be made based on a balanced consideration of the impact on producibility, reliability, and maintainability due to increased system complexity as well as cost. D. Weight Dierential Equation

The most widely used mission analysis technique for conventional aircraft is based on the assumption that the rate of change in aircraft weight equals the fuel ow. However, aircraft such as the Helios, which is equipped with regenerative power systems, maintains the same weight during the entire mission. Furthermore, more stringent emissions regulations of the future may force aerospace engineers to innovate propulsion systems so that the system can separate specic by-product components from engine emissions and store them onboard during ight. In order to capture such a characteristic, a more generalized weight decomposition has been developed as follows: W = WE + WP + WCE + WR (13) where WR is the weight of the retained products. The non-consumable energy weight is included in the empty weight. Taking the derivative of Eq. (13) with respect to time produces the rate of change in aircraft weight as W = WCE + WR (14) where the dotted () notation represents dierentiation with respect to time. By assuming that the weight of the retained products is proportional to fuel consumption, WR = WCE where is the stored-product-to-fuel ratio. Substituting Eq. (15) into Eq. (14) yields W = (1 )WCE (16) (15)

By introducing a constant k, the relationship between the rate of change in the aircraft weight and the rate of change in the consumable energy weight is given as W = k WCE where k =1 (18) The value of k is determined by the characteristics of the involved energy conversion processes. In the case of traditional IC engines, k equals one. In the case of zero-emission aircraft that breathes external oxygen to burn hydrogen and store water onboard,10 the aircraft weight will grow as fuel burns, and thus, k is a negative number. If the aircraft is powered by multiple power paths, k is given as
nCE

(17)

k=
i=1

1
nCE ii=1
ii i

i CE
ii i i

(19)

CE

ii

IV.

Formulation Part I: Constraint Analysis

Application of Newtons second law to aircraft motion yields (T (D + R)) V = W dh W d + dt go dt V2 2 (20)

where T is thrust; D is basic conguration drag; R is additional drag to D due to the changes of the conguration such as deecting control surfaces, extracting landing gear, and installing ordnance such as
8 of 18 American Institute of Aeronautics and Astronautics

weapons and external fuel tanks; h is the ight altitude; V is the free-stream velocity; and go is the gravity constant. Starting with Eq. (20), Mattingly derived the general form of the constraint equations that provide the desired relationships between the thrust-to-weight ratio and wing loading as follows:20 TSL = WTO qS K1 WTO n WTO q S
2

+ K2

n WTO q S

+ CDo +

R 1 d + qS V dt

h+

V2 2go

(21)

where is the weight fraction; is the thrust lapse; q is the dynamic pressure; K1 is the drag polar coecient for the 2nd order term; K2 is the drag polar coecient for the 1st order term; CDo is the zero lift drag coecient; and n is the load factor. By replacing T with P/V , this equation can be reformed to a power-based equation as follows P = WTO qS K1 WTO n WTO q S
2

+ K2

n WTO q S

+ CDo +

R 1 d + qS V dt

h+

V2 2go

(22)

When Eq. (10) is substituted into Eq. (22) to signify dierent power paths, a new constraint equation can be formed as follows. poSL i = i WT O i
i

qS K1 WT O

n WTO q S

+ K2

n WTO q S

+ CDo +

R 1 d + qS V dt

h+

V2 2go

(23)

If the appropriate assumptions for each performance requirement are applied to this master equation, a corresponding constraint equation can be derived in reduced form. Readers can nd a set of constraint equations in a previous paper17 of the authors. The weight fraction values for each mission segment are not 250.0 available at this moment. Therefore, these values must be reasonably assumed for this analysis and should be updated by 200.0 b a the mission analysis, if necessary. Once the constraint equa150.0 tions are developed, the feasible solution area can be graphically identied as bounded by constraint curves. The method of 100.0 selecting the design point is similar to the traditional method, which is discussed in detail in Reference [20]. The only dier50.0 ence is that the power-to-weight ratio instead of the thrust-to0.0 weight ratio is selected. 0 10 20 If the aircraft is powered by multiple power paths, however, 50.0 Take-off Roll as many constraint analysis domains as power paths exist, so Cruise 45.0 Landing Roll the design point selection process becomes more complicated. 40.0 Climb Approach 35.0 An example of the constraint analysis for a multiple-powerDesign Point 30.0 path propulsion system is illustrated in Figure 4. The propul25.0 sion system consists of the primary power path that provides 20.0 power during the entire mission and the secondary power path 15.0 10.0 that augments the primary power on take-o, climb, missed 5.0 approaches, and emergency situations. The design point of the b a 0.0 primary power path is selected at the location of the lowest 0 10 20 Wing Loading power-to-weight ratio. Then, the design point of the secondary power path must be determined at a location along the ver- Figure 4. Design point selection of multitical line, a-a, because wing loading of the aircraft must be power path maintained through all the constraint analysis domains. If a dierent wing loading on line b-b is selected, the power-toweight ratio of the primary power path increases, while the power-to-weight ratio of the primary power path decreases. Another possible trade-o is to change the power contribution ratio of the two power paths. The constraint curves of both the primary power path and the secondary power path are functions of the power distribution ratio between the two power paths. As the contribution ratio of the secondary power path increases, the primary power path is allowed to have a lower power-to-weight ratio. Therefore, determination of the best sets of design points simply based on the constraint analysis such as this is not always easy. The best set of design points may be found by combining the mission analysis and employing an optimization process.
Primary Power Path Secondary Power Path

9 of 18 American Institute of Aeronautics and Astronautics

Po/W

Po/W

V.

Formulation Part II: Mission Analysis

The mission analysis of the proposed method is similar to that of the traditional method. The mission prole is chopped into small segments, which allows us to assume that several parameters are constant in each segment and simplify the associated equations a great deal. Then, for each segment, the weight fraction or the amount of both consumable energy and non-consumable energy is calculated. By summing the results, the amount of energy required to perform an entire mission can be estimated. A. Consumable Energy Sizing
sth Segment

Consumed energy is power multiplied by time:


i dECE

poi

dt =

P
i

dt

(24)
Climb 1st Segment

Cruise s-1 s Descend

The change in the amount of consumable energy is proportional to the change in the weight of the energy sources. dECE = CE dWCE Combining Eq. (24) and Eq. (25) yields
i dWCE i i i

Loiter n Landing

(25)

0 1 Warm up Take-off and Taxi

Payload drop

P
i

Figure 5. Mission Prole Segments

CE

dt

(26)

where CE is the specic energy of the fuel in the power path. As mentioned before, two types of energy are being considered. The ways of calculating the amount of consumable energy dier, depending on how aircraft weight varies. 1. Variable Aircraft Weight (k = 0)

If aircraft weight changes in the mission segment, the amount of consumable energy consumed in the mission segment can be calculated as follows. Substituting Eq. (26) into Eq. (17) yields
CE dW k i = i i W i=1 CE

P W

dt

(27)

By solving Eq. (27) for the sth mission segment, the weight fraction of the segment obtained is W (s) = exp W (s1) If the segment is small enough such that k equations can be approximated as follows:
t(s) t(s1) i nCE

i i i=1 CE
i

P W

dt

(28)

/CE

can be assumed to be constant, then the above

W (s) (s) = exp k (s) (s) CE W (s1) where (s) is the specic mechanical energy, which is given as
(s) (s) t(s)

(29)

=
t(s1)

P W

dt

(30)

and CE is the overall power specic fuel consumption (OPSFC), which is equivalent to the power-specic fuel consumption (PSFC) of conventional combustion engines and given as CE =
i=1 (s) nCE

i i

CE

(31)

10 of 18 American Institute of Aeronautics and Astronautics

The ways of computing (s) dier, depending on the existence of excess power. In the case of positive excess power, P d(h + V 2 /2go ) dze dt = = (32) W 1u 1u where nT i i D+R i PoSL CD + CDR u= / = nV (33) T CL WTO i=1 Therefore, Eq. (30) can be rewritten as
(s)

(s) ze

ze

(s1)

dze 1u

ze 1u

(s)

(34)

In the case of zero excess power, such that during cruise or a sustained turn, required power equals the total drag multiplied by the free-stream velocity as follows: P dt = W or P dt = W D+R W D+R W V dt (35)

ds

(36)

Therefore, Eq. (30) can be rewritten as (s) = or


(s) t(s) t(s1) s(s)

D+R W D+R W

V dt

CD + CDR CL CD + CDR CL

V t

(37)

=
s(s1)

ds

(38)

Then, the ratio of the consumable energy weight used in the mission segment can be expressed as follows: WCE W (s1) W (s) (s1) = = WTO kWTO k where (s1) = When payload that weights WP
(s1) (s)

W (s) W (s1)

(39)

W (s1) WTO

(40)

is dropped at the beginning of the segment, (s1) is given as (s1) = W (s1) WP WTO

(s1)

(41)

where W (s1) is the weight of the aircraft right before the payload drop. 2. Constant Aircraft Weight (k = 0)

If aircraft weight does not change in the mission segment, and the propulsion system consumes a certain type of fuel, fuel consumption is in proportion to power consumption. By dividing Eq. (26) by the aircraft weight, nCE dWCE i P = dt (42) i i W W i=1
CE

Note that W is constant. By integrating equation Eq. (42) into the ight time or the ight distance of the sth mission segment, (s) t(s) nCE WCE i P = dt (43) i i W W (s1) t i=1 CE
11 of 18 American Institute of Aeronautics and Astronautics

or

WCE (s) = (s1) (s) CE (44) WTO The weight of the total consumable energy normalized by the take-o gross weight, CE , can be obtained as the sum of the normalized weight of the consumable energy of the individual segment. CE = WCE = (1 + WTO
m CE ) s=1

(s)

WCE WTO

(s)

(45)

Where CE is the consumable energy allowance ratio that accounts for an propulsion system with poor-thannormal energy consumption and the amount of unusable energy. In the traditional aircraft sizing method, the total aircraft fuel generally includes the mission fuel as well as an 5% allowance for the reserve fuel which accounts for an engine with poor-than-normal fuel consumption and additional 1% allowance for trapped fuel. Therefore, CE for the conventional aircraft is 0.06. However, the proper allowance for unconventionally powered aircraft will dier, depending on the characteristics of the propulsion system and the energy storage devices. B. Non-consumable Energy Sizing

Consumed energy is power multiplied by time: dENE = poj dt = Equation (46) can be rewritten as dENE j P = j dt WTO W The amount of the j th energy consumed in the sth segment can be expressed as ENE WTO
j (s1) j j

dt

(46)

(47)

ENE WTO

(s)

t(s)

=
t(s1)

(s1)
j

P dt W

(48)

Total non-consumable energy can be calculated by summation of the consumed energy of all the segments. (0) (1) (1) (2) j j j j ENE ENE ENE ENE + + + WTO WTO WTO WTO (s1) (s) (m1) (m) j j j j ENE ENE ENE ENE + + = (49) WTO WTO WTO WTO ENE WTO Therefore,
j j (0)

ENE WTO

(m)

ENE WTO
m

ENE = (s) WTO s=1 and WNE =


j=1 nNE j

(s1)
j nNE j=1

(50)
j

WNE =

ENE NE
j

(51)

The weight of the total consumable energy normalized by the take-o gross weight, CE , can be obtained as follows: m WNE (s) = (1 + NE ) (s1) (s) NE NE = (1 + NE ) (52) WTO s=1
12 of 18 American Institute of Aeronautics and Astronautics

Where NE is the non-consumable energy allowance ratio. Once this analysis is complete, the assumed weight fraction, , at each mission segment in the constraint analysis must be evaluated with the results from the weight fraction analysis. If the initial assumption on turns out to be inaccurate, the processes described up to this point may have to be iterated.

VI.

Formulation Part III: Weight Estimation

Equation (13) can be rewritten as follows at take-o: WTO = WE + WP + WCE (53)

The payload weight, WP , is usually given as a part of the customer requirements, and the consumable energy weight is determined by Eq. (45). The traditional ways of estimating the empty weight dier, depending on how much data are available. The rst choice is to use a simple relationship between the take-o gross weight and the empty weight. If we have more information, we may then be able to calculate the weight of each group or the weight of the components. Either way, the empty weight can be formed into an empty weight fraction, , multiplied by the take-o gross weight as follows: WE = WTO (54)

However, the empirical empty weight fraction, , is not available for revolutionary concepts. In order to obtain a relationship between the empty weight and the take-o gross weight, required in the proposed formulation, the empty weight is broken into subgroup weight terms as follows: WE = WE + WPD + WNE + WE (55)

where WE is the empty weight less the weight of installed propulsion systems; WPD is the weight of power devices; and WE is the weight correction to the traditional empty weight. Then, WE = WTO + WTO + NE WTO + WTO where = WPD WTO (56)

(57)

By combining Eq. (45), Eq. (53), and Eq. (56), the take-o gloss weight equation is given as follows: WTO = WP NE CE 1 (58)

The take-o gross weight equation, Eq. (58), cannot be solved in a closed form, since , , and are also functions of the take-o gross weight. Therefore, an iterative process must be used to solve the equations.

VII.

Sizing Example

In order to demonstrate the proposed method, the formulation was applied to a Global Hawk-like HALE conguration powered by an all-electric propulsion system. The sizing mission prole of the aircraft is illustrated in Figure 6. The aircraft performs surveillance missions for 24 hours at a station located about 3,000 nm away from the base. It carries 1,900 lbs. of payload, including synthetic aperture radar, a digital charge-coupled device (CCD) camera, and a third-generation infrared sensor system. The mission also requires high-power extraction (40 Hp at 65,000 feet) to drive the electric payloads.23 The general arrangement of the baseline conguration and the schematic diagram of the electric propulsion system are illustrated in Figure 7. The conguration is featured by a high aspect-ratio wing and V-tail arrangement. The aerodynamic characteristics of the conguration were analyzed by the Boeing Design and Analysis Program (BDAP),24 developed by Boeing for calculating for zero-lift drag, and a generalized vortex lattice program, VORLAX,25 developed by Lockheed for calculating the induced drag.

13 of 18 American Institute of Aeronautics and Astronautics

The Global Hawk is equipped with a single turbo-fan 24 hr Altitude engine on top of the after-fuselage. The electric HALE loiter 65K ft aircraft is powered by multiple propellers driven by elec- 65K ft Egress tric motors mounted on the bottom of the wing. The elec- 50K ft Climb Cruise 50K ft Climb 3000 Nm tric propulsion system consists of proton exchange mem3000 Nm brane (PEM) fuel cells, a power management and distribution system (PMAD), main motors, and other accesIdle/Takeoff Descent/Land sories, as illustrated in Figure 7. Two dierent levels of notional technology for the propulsion system, state-ofFigure 6. Mission prole the-art technology and advanced technology, which are characterized by the data listed in Table 1, are applied to this sample study. The eciency of the propeller is estimated using GTPROP, in-house code originally developed and validated by Hamilton Standard, and the weight of the propellers is estimated by employing an empirical equation in the reference.19 The weight of the state-of-the-art motors is calculated based on data collected from electric car companies. The weight of super conducting motors and crycoolers is calculated from a tool obtained from a cooperating research group at Florida A & M University. The weight of other accessories are estimated by using a simple linear relation based on data from a NASA technical report.8

H2

Hydrogen Tank

Hum

Air

C*

Motor

PEM Fuel Cell

Air, H2O

Motor

PMAD Batteries Aux HX

* Compressor

Figure 7. Baseline conguration and electric propulsion system.

Fuel Cells PMAD Main Motors

Efficiency (%) Specific Power Specific Power (hp/lb) Type Efficiency @ Max. Continuous Power Specific Power (hp/lb)

Off-the-shelf Advanced Technology 45.7 Increased by 10% 0.485 Increased by 50% 0.65 0.9 Blussless DC Motor High Temperature Super Conducting Motor 93 98.7 0.7 Increased by approximately 5 times

Table 1. Two dierent levels of technology.

The propulsion system uses hydrogen as fuel. The gaseous hydrogen fuel can be stored in the following ways: pressure tank, metal hydride, carbon nanotubes, and glass microspheres are methods for storing gaseous hydrogen. A cryogenic tanker is used for gelled or liquid hydrogen.26 Liquid hydrogen was chosen for this study because the previous research12, 16, 26 proved it to be the best t to applications of aircraft. A model that estimate the weight of the liquid hydrogen tank based on the formulation in Chambliss and Kelly27 was developed by a co-researcher and was used for this study. The results of the constraint analyses of the electric HALE aircraft with two sets of technologies are presented in Figure 8. The design point for the aircraft with the o-the-shelf technology is selected as 40 lbs/ft2 of wing loading and 57 Hp/lbs of power-to-weight ratio. Wing loading of the aircraft with the advanced technology remains the same and the power-to-weight ratio decreases to 55 Hp/lbs, thanks to the benets of increased eciency of fuel cells and the motors. With these combinations of wing loading and power-to-weight ratio, the mission performance for two technology sets are analyzed with 6% of energy
14 of 18 American Institute of Aeronautics and Astronautics

Table 2. Sizing results

Wing Area (sq. ft) Take-o Gross Weight (lbs) Total Shaft Power (Hp) or Thrust*(lbs)

Global Hawk 540 25,600 8,600*

O-the-shelf 2,086 83,435 3,671

Advanced Technology 521 20,821 921

Off-the-shelf Technology

Advanced Technology

Po SL = 57 W TO

W TO = 40 S

Po SL = 55 W TO

Figure 8. Constraint analyses of the electric HALE conguration with two sets of technologies.

Weight Breakdowns of Propulsion Systems


90000 80000 70000 60000 Weight (lbs) 50000 40000 30000 20000 10000 0 Global Hawk Off-the-s helf Advanced Technology
Compressor, 245, 7% Humidifier and Intercooler, 326, 9% PMAD, 1037, 28% PMAD, 6073, 26% Prop, 1478, 6% Motor, 5245, 22%

Payload Wf Airfram e and Subs ys tem s H2 Tanker Propuls ion

Compressor, 1036, 4% Humidifier and Intercooler, 1380, 6%

Fuel Cell, 8140, 36%

Off-the-shelf
Prop, 431, 12% Motor, 350, 10%

Fuel Cell, 1283, 34%

Advanced Technology

Figure 9. Comparison of weight breakdowns.

15 of 18 American Institute of Aeronautics and Astronautics

  A C  B 4& 8 @9 2 87 50 6 1 &432 1 0 (('&%$ 0) 


W TO = 40 S

    

  " # ! E S E H E R

VU W ` V Uxp ` Y v yt Uu pxUwv u t` rrqpiYh ts ` WVa Y WVT U`X U E E G Q E D E P

EF EE D EF G E EF H E EF I E

EF E c EP b d f EF D g EP e EF G EP EF H EP EF I EP EF E ED

allowance. The results from sizing the electric HALE aircraft with two sets of technologies are compared in Table 2. The sized aircraft with o-the-shelf technology is approximately four times as large and three times as heavy as the Global Hawk. Since the aircraft weight is far beyond the reliable range of the regression equation, these weight values are not reliable. However, the results provide sucient information to prove that the o-the-shelf technology has not yet matured enough to the point powering high-speed HALE aircraft. With the infusion of a set of advanced technology, the weight and wing area of the aircraft are signicantly reduced. The weight breakdown of the three aircraft are compared in Figure 9. The most notable observation is that the fuel fractions of both electric HALE aircraft signicantly decrease while the ratio of the propulsion system weight greatly increases, which indicates that the fuel cell propulsion system is an ecient, yet heavy system.

VIII.

Comprehensive Sizing Method

The proposed method mostly concerns balancing power and energy. However, a successfully sized conguration must achieve one more criterion: the balance between the required aircraft volume and the available aircraft volume, as illustrated in Figure 10. In general, the volume balance is veried through more detailed studies of the internal arrangement after the initial aircraft conguration is fully established through the aircraft sizing process. In the case of traditional aircraft design, however, the volume balance is implicitly secured to a certain degree via the application of historical regression rules to weight estimation without direct assessment of volume balance, simply because all existing aircraft whose weight data are used to construct the regressed equations contain all subsystems, structures, and fuel inside the aircraft. In addition, it is not too far-fetched to regard the aircraft being designed as a small perturbation from the historical trend. Nevertheless, in the case of designing a revolutionary aircraft that uses unconventional propulsion systems and consumes unconventional energy sources, such an implicit volume balance will not work. For instance, the converged congurations in the sample sizing study, where volume balance was not considered, may not provide sucient volume for all required systems. If they do not, the external conguration may be modied so that the eciency of internal packaging increases, or the aircraft may need to be scaled up beyond the minimum size at which both power balance and energy balance are achieved. Therefore, early consideration of the impact of unconventional energy and/or the propulsion system on the required volume will be essential so that rework due to volume issues can be avoided in the following design phase. All three criteria will not always play equal roles in the aircraft sizing process. The relative importance among the three depends on various parameters such as sizing requirements, the shape of the conguration, and the characteristics of the propulsion systems and energy sources, including the following four parameters: the specic energy and energy density of the energy source, and the specic power and power density of the propulsion system. For instance, if the aircraft is powered by an isomer energy source whose specic energy is exceptionably higher than that of conventional fuel (approximately 3 104 times greater), the weight of the energy source may be trivial. Therefore, energy balance will not be a major concern in the aircraft sizing process any longer, just as the weight of the cockpit instruments is not worried much in a conventional aircraft sizing practice.

IX.

Conclusion

This paper has discussed the need to develop a generalized sizing method that is applicable to unconventionally powered aircraft. In order to ensure application to a wide range of unconventional energy sources and revolutionary propulsion systems, the new method must embrace key modications to traditional methods. The modication was built upon generalizations made from the following three parts of the traditional aircraft sizing method. First, the propulsion system is modeled as an integration of multiple power paths, each of which is characterized according to three parameters: the specic energy of the energy source, the specic power and the eciency maps of the power transfer devices. Therefore, this method can capture the key characteristics of any new propulsion system and incorporate them into the aircraft sizing process. Secondly, fuel is generalized as a concept of onboard energy that can originate from a variety of energy sources. Each energy source is categorized as either consumable or non-consumable. Finally, a more generalized weight decomposition equation and weight dierential equation are employed to model the fuel consumption behavior of the revolutionary concepts that, unlike conventional ICs, may retain specic by-products from engine emissions onboard during ight.

16 of 18 American Institute of Aeronautics and Astronautics

Point Performance Requirements

Aerodynamics
Constraints Analysis

Weight Estimation

Propulsion

Mission Analysis

Weight Notional Concept


Mission Performance Requirements A/C Scaling Factor and Subsystem Weight

Volume Analysis

Volume

Figure 10. Comprehensive sizing method.

This paper has also discussed a more comprehensive sizing method. This method includes volume balance, which is one of three criteria to be considered in the aircraft sizing process, but which is openly overlooked in the sizing conventional aircraft. This sizing method will allow system design engineers to evaluate the impact of new propulsion concepts and alternate energy sources on aircraft integration. In addition, the capability of this method will be enhanced to allow further investigation of optimum congurations for revolutionary technologies by combining optimization techniques.

Acknowledgments
The authors would like to thank NASA and the Department of Defense for supporting this research under the URETI for Aeropropulsion and Power Technology (UAPT). In addition, we would like to thank our co-workers in the URETI project, particularly Mr. Choi for supporting the modeling and analysis of the electric propulsion system, Mr. Agrawal for providing a tool that estimated the hydrogen storage device weight, and Dr. Masson at FAMU for providing a design and analysis tool of a super-conducting motor for this research.

References
1 Martinez, J. S. and Plebuch, R. K., Nuclear Propulsion Applications, American Inst. Of Aeronautics And Astronautics, Annual Meeting And Technical Display, 4th, No. AIAA-1967-781, 1967. 2 Review of Manned Aircraft Nuclear Propulsion Program, Atomic Energy Commission and the Department of Defense, 1963. 3 Beamed Laser Power For UAVs, NASA Facts, Dryden Flight Research Center, March 2004, FS-2004-03-087 DFRC. 4 Hertzberg, A. and Sunt, K. C., A laser-powered ight transportation system, American Institute of Aeronautics and Astronautics, Aircraft Systemsand Technology Conference, Los Angeles, Calif., August 21-23, 1978 , 1978. 5 Hamilton, C. E., King, P. I., and Franke, M. E., Isomer Energy Source in Hybrid Jet Engines for High Altitude Reconnaissance Flight, Journal of Aircraft 2004 , Vol. 41, No. 1, 2004, pp. 151155. 6 Frisbee, R. H., Advanced Space Propulsion for the 21st Century 0748-4658, Journal of Propulsion and Power , Vol. 19, No. 6, 2003, pp. 11291154.

17 of 18 American Institute of Aeronautics and Astronautics

7 Bulman, M. J. and Borowski, S., Deep Space Propulsion Requirements Development, 39th AIAA/ASME/SAE/ASEE Joint Propulsion Conference and Exhibit, Huntsville, Alabama, July 20-23 , No. AIAA-2003-5129, 2003. 8 Freeh, J. E., Liang, A. D., Berton, J. J., and Wickenheiser, T. J., Electrical Systems Analysis at NASA Glenn Research Center: Status and Prospects, Tech. rep., NASA Glenn Research Center, 2003. 9 Guynn, M. D. and Olson, E. D., Evaluation of an Aircraft Concept With Over-Wing, Hydrogen-Fueled Engines for Reduced Noise and Emissions, Tech. rep., NASA Langley Research Center, Hampton, Virginia, 2002. 10 Guynn, M. D., Freeh, J. E., and Olson, E. D., Evaluation of a Hydrogen Fuel Cell Powered Blended-Wing-Body Aircraft Concept for Reduced Noise and Emissions, Tech. Rep. NASA/TM-2004-212989, NASA, 2004. 11 Vehicle Systems Program Selected Projects, NASA Aeronautics News, Vol. 5, 2004. 12 Wickenheiser, T. J., Sehra, A. K., Seng, G. T., Freeh, J. E., and Berton, J. J., Emissionless Aircraft: Requirements and challenges, AIAA/CAS International Air and Space Symposium and Exposition: The next 100 Y 14-17 July 2003, Dayton, Ohio, No. AIAA 2003-2810, July 2003. 13 Alexander, D. S., Advanced Energetics for Aeronautical Applications, Tech. Rep. NASA/CR-2003-212169, MSE Technology Applications, Inc., 2003. 14 Berton, J. J., Freeh, J. E., and Wickenheiser, T. J., An Analytical Performance Assessment of a Fuel Cell-Powered, Small Electric Airplane, Tech. Rep. NASA/TM?003-212393, NASA Glenn Research Center, June 2003. 15 Kohout, L. L. and Schmitz, P. C., Fuel Cell Propulsion Systems for an All-Electric Personal Air Vehicle, Tech. Rep. NASA/TM?003-212354, NASA Glenn Research Center, July 2003. 16 Alexander, D., Lee, Y.-M., Guynn, M., and Bushnell, D., Emissionless Aircraft Study, 38th AIAA/ASME/SAE/ASEE Joint Propulsion Conference & Exhibit, No. AIAA-2002-4056, 2002. 17 Nam, T., Soban, D. S., and Mavris, D. N., Power Based Sizing Method for Aircraft Consuming Unconventional Energy, 43rd AIAA Aerospace Sciences Meeting and Exhibit, Reno, Nevada, Jan. 10-13, 2005 , No. AIAA-2005-818. 18 Raymer, D. P., Aircraft Design: A Conceptual Approach, AIAA, 1999. 19 Roskam, J., Airplane Design, Roskam Aviation and Engineering Corp., 1985. 20 Mattingly, J. D., Heiser, W. H., and Daley, D. H., Aircraft Engine Design, AIAA Education Series, sixth ed., 1987. 21 Holder, B. and Wallace, M., Range Unlimited - A History of Areial Refueling, A Schier Military History Book, 2000. 22 Bayles, G., New Power Source Technologies for Electric Wheelchairs, Tech. rep., University of Pittsburgh RERC on Wheelchair Mobility, 9 1996. 23 Schelp, T., Corea, V., and Jeries, J., Development of the RQ-4A Global Hawk Propulsion System, 39th AIAA/ASME/SAE/ASEE Joint Propulsion Conference and Exhibit, Huntsville, Alabama, July 20-23, 2003 , No. AIAA-20034680, 2003. 24 Middleton, W. D. and Lundry, J. L., A System of Aerodynamic Design and Analysis of Supersonic Aircraft, Tech. Rep. NASA CR-3351, 1980. 25 Miranda, R. L., Elliott, D. R., and Baker, M. W., A Generalized Vortex Lattice Method for Subsonic and Supersonic Flow Applications, Tech. Rep. NASA CR-2865, 1977. 26 Colozza, A. J., Hydrogen Storage for Aircraft Applications Overview, Tech. Rep. NASA/CR?002-211867, Analex Corporation, September 2002. 27 Chambliss, K., Kelly, S., and Kimble, J., Cryogenic Fluid Storage For The Mission To Mars, Tech. rep., Texas Tech University, 1999.

18 of 18 American Institute of Aeronautics and Astronautics

You might also like