A Function-based Strategy for Analysis of Energy Systems in Transportation Vehicles

Six functions are identified as the most critical (“core” functions) to transportation vehicle energy systems. These selections are validated through analysis of 25 function structures as well as observations of a number of existing energy systems. Identifying which of the core functions and which of the energy types are involved in a given energy system is the Core-Function Modeling strategy (CFM strategy). These functions and energy types (the framework of CFM) are used to categorize approximately fifty processes and devices. This list is the Energy Morph Matrix (EMM). An experiment is performed that demonstrates how the EMM can be used as an aid to the concept generation process. The EMM also adapts well to a more automated approach for designing energy systems when used in combination with a search algorithm to identify chains of energy components. By incorporating a metric such as system efficiency or energy density into the search and computing this metric for each chain of energy components, these chains can be ranked and leading candidates can be highlighted for further analysis. INTRODUCTION The realm of all existing energy-related devices and processes is incredibly vast. Even when analysis is restricted to the transportation sector, the number of components and devices that use or convert energy remains unmanageably large. From a design standpoint, choosing the most appropriate component or device for a particular purpose among the vast number of alternatives presents a challenge. For example, consider the relatively simple hypothetical task of designing the best possible actuator for the control surface of a micro aerial vehicle (MAV). How should this problem be solved? Servo motors are very frequently used for this application because they accomplish the desired functionality with tolerable performance. But are servo motors really the best type of actuator for all types of MAVs, operating environments, and size scales? To answer this question, servo motors must be compared to other devices that could accomplish the same function. Ideally, there would be some “master list” of actuators for MAVs, and servo motors could be exhaustively compared to everything on the list to prove that they are the best. Such a list would be convenient for this specific design problem, but it would be impossible to have a list for every single design problem for energy systems. Fortunately, patterns do emerge when considering the entire realm of energy-related devices and processes, and these patterns can be used to make design processes more effective. For an example of such a pattern, note that actuators typically convert some form of energy into mechanical energy. Rather than list and compare every type of actuator for MAVs, it is Proceedings of the ASME 2009 International Design Engine ring Technical Conferences & Computers and Information in Engine ring Conference IDETC/CIE 20 9 August 30 September 2, 20 9, San Diego, California, USA This work is in part a work of the U.S. Government. ASME disclaims all interest in the U.S. Government's contributions. Copyright © 2009 by ASME Downloaded 25 Aug 2012 to 128.83.63.20. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm 2 Copyright © 2009 by ASME potentially more useful (and more applicable to other design problems) to list and compare devices that perform this type of energy conversion. This is an example of a function-based approach to the design process. Functional analysis is a branch of design theory that includes the understanding or conceptualization of an artifact as a conglomeration of functional subunits. It has been tested and shown to be a useful tool in engineering design [1-5]. Describing an actuator functionally—as a converter of energy from the electrical to the mechanical domain—allows it to be more easily compared to other devices which accomplish similar functionality. A variety of other materials and devices can be considered as potentially useful actuators—such as shape-memory alloys, which convert heat energy into mechanical energy, or piezoelectric materials, which convert electrical energy into mechanical energy (and do so with substantially fewer parts than servo motors). Energy systems are not typically comprised of single components, but rather combinations of components and processes working together. Also, energy systems are comprised of components that perform a variety of functions. To reconcile a function-based design approach with the large number of energy system components and component functions, a morphological analysis approach to energy system analysis and design is proposed. Morphological analysis is a well-established design tool [2,3,6] and involves listing phenomena, concepts, ideas, or physical components along two or more axes that relate in some way to a problem to be solved. Then, the total set of possible relationships or configurations contained in a problem can be systematically investigated [7]. Design and analysis of energy systems in transportation vehicles can be made easier using morphological analysis if suitable “axes” can be developed. This paper proposes three energy-related functions and six energy domains as axes for morphological analysis. By taking a function-based morphological analysis approach to the design and analysis of energy systems, a large number of potentially viable energy systems can be rapidly identified, and the inherent patterns in the energy landscape can be more readily understood. This approach is also highly adaptable to more automated design methods where computer tools can be used to identify viable energy systems. Throughout this paper, micro aerial vehicles (MAVs) will be used as illustrative vehicle examples to help explain concepts. A number of the ideas stated in this paper were developed with MAVs in mind, and then generalized to include all transportation vehicles. Research Goals There are two main goals of this research in energy systems. Prior to this work, there has been no standardized or universally adopted way to catalogue or organize devices and processes that pertain to energy. For this reason, there is no single list of energy-related devices and processes for energy system designers to consult or utilize. The first goal of this research is to devise a logical way based on functional modeling to organize and describe the design space for energy systems in transportation. Since this framework is based on the most critical or most frequently performed functions pertaining to energy systems (the “core” functions for energy systems), it is titled the Core-Function Modeling (CFM) strategy. The second goal is to create an extensive example list of devices and processes according to CFM strategy (this is the Energy Morph Matrix or EMM) and then comment on and demonstrate its implications for design automation and design processes in general. Research Approach A combination of inductive and deductive research approaches is used to investigate the energy system design space. The core functions are initially identified by empirical observation of existing systems and the patterns and commonalities arising among them. This inductive approach starts with specific real-world examples and postulates more general rules about the design space of energy systems. The selection of these core functions is then validated by a more deductive approach based on functional analysis. This validation begins with a basis set of 42 functions [8]. Next, functions that do not apply to energy are pared off the list. By analyzing existing function structures, the incidence of each of these functions that apply to energy is studied. The most frequently occurring functions are compared to the core functions identified by the empirical observation phase of research. The core functions are then used to categorize a large number of existing devices and processes by function and by energy domain. This categorized list (the EMM) is used in a concept generation experiment in order to assess its viability as a design tool, and a plan is proposed describing one way the EMM could be used to facilitate increased design automation. IDENTIFYING CORE FUNCTIONS BY OBSERVATION In the following discussion, it is assumed that certain functions exist that, taken together, span the space of all energy system components. The existence of this smaller subset of all functions has not been proven, but the following discussion provides evidence that it could exist. In the next sections, each proposed core function is introduced, followed by empirical observations that support its inclusion in the set of core energy functions. These observations also indicate that the proposed set of core energy functions appears to span the energy domain. Import Transportation vehicles, by definition, transport people or goods (in unmanned vehicles, these “goods” are often sensors) from one place to another. Any object with mass that is in motion possesses kinetic energy. Due to the law of energy conservation, the kinetic energy possessed by the vehicle in motion must be obtained from some source—from within the vehicle’s control volume or outside of it. Imported energy can Copyright © 2009 by ASME Downloaded 25 Aug 2012 to 128.83.63.20. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm 3 Copyright © 2009 by ASME take many different forms. Energy can be imported prior to a vehicle’s journey in the form of stored energy—such as liquid fuel or charged batteries. Energy can be imported during a vehicle’s journey—as microwave power sent wirelessly or perhaps in the form of mechanical energy simply pushing the vehicle along. Energy is constantly flowing into and out of every realworld vehicle’s control volume. Energy flow into the vehicle is imported energy, and energy flow out of the vehicle is exported energy. For a practical example, consider the case of a gaspowered MAV in flight. At any given moment, energy is being imported into the MAV in myriad forms—inbound radio communications exciting antennae, sunlight warming the MAV’s surfaces, or a tailwind propelling the MAV forward with mechanical energy. Export Continuing with the example of the MAV in flight, consider the concept of a closed system. A closed system is completely isolated from its surroundings, and exists exclusively in the realm of theory because in reality, every system interacts in some way with its surroundings. At all times, both matter and energy are being exchanged between the MAV and its surroundings. Energy is being exported in the form of hot exhaust fumes from the gas-burning engine, momentum and heat of the air itself displaced by the moving aircraft, outbound radio communications, sound waves, and radiant heat from the hot engine compartment. “Export” must be another core function because of the constant energy exchange that exists between all systems and their surroundings in the real world. Exported energy is a useful concept to consider because often a large percentage of energy that is exported is wasted energy. By noting what energy is being exported, opportunities to recapture or reuse this energy can be found. Store Although not all transportation vehicles have the ability to store energy, the vast majority do store energy in some form. Every commercially available automobile stores energy, mostly in the form of gasoline. MAVs typically store energy in the form of batteries, alcohol-based liquid fuel, or petroleum-based liquid fuel. To be entirely portable and to move about untethered from any fixed object as most transportation vehicles do, energy must be carried by the vehicle itself or harvested from the environment. As energy is required to move through air, and since this energy is rarely imported into the vehicle while it is traveling, energy must be stored on the vehicle. The issue of stored energy is of paramount importance when transportation vehicle range is concerned. Simply put, vehicles that can store more energy (per added fuel weight) can travel farther without refueling. Although much engineering effort is also expended in increasing vehicle energy efficiency and reducing the need for energy, storage media high in energy density remain in demand. Convert The “convert” function, in which energy is converted from one form to another, occurs very frequently within all consumer products and transportation vehicle systems. Energy exists in a variety of forms—thermal, chemical, mechanical, etc.—and in each of these forms, energy can be used to accomplish a variety of ends. For example, thermal energy can warm people in cold climates or it can be used to facilitate chemical reactions in manufacturing processes. Mechanical energy can move people and goods from place to place or it can be used to cut materials to different shapes and sizes. Each form of energy has both obvious uses and obvious limitations. Electrical energy is incredibly easy to transmit long distances since it requires only a conductive wire to travel at near-light speed, but by itself has limited usefulness. Once the electrical energy reaches its destination, it is readily converted into other more useful forms. Large amounts of chemical energy can be stored in the form of gasoline, but only by being converted into heat and subesquently mechanical energy can it be used to propel a vehicle. Transport Clearly supported by intuition as well as observation, the movement of energy from one place to another is a critical function for all systems that use energy. Within the confines of an MAV, transport of electrical energy, although important, might only amount to a few feet of wire. Other forms of energy transmission crucial to MAV functionality are long distance data transmissions and energy sensing (sensing radiant energy in the form of visible or infrared light, for example). Also, some UAVs are capable of utilizing energy transported wirelessly from the sun, or from microwaves emitted from a ground station. Regulate Effective regulation of energy can often be the difference between a working system and one that malfunctions. A clear example of the importance of effective energy regulation can be found under the hood of any automobile—fuel injection systems and the carbureted systems they replaced are responsible for metering out a precise amount of liquid fuel (chemical energy) so that the engine runs efficiently. “Regulate” is another likely core function because it is critically important to the functioning of nearly every existing energy system. VALIDATING CORE FUNCTIONS Starting Point: a Functional Basis The functional basis [8,9,10] developed and advanced by Stone and Wood (2000) serves as a suitable starting point for a function-related method of organizing and categorizing the many different elements that make up the whole realm of devices and processes pertaining to energy management and utilization. Stone and Wood reconciled and integrated research efforts by a number of other teams into a standardized set of function-related terminology, culminating in a basis set of 42 Copyright © 2009 by ASME Downloaded 25 Aug 2012 to 128.83.63.20. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm 4 Copyright © 2009 by ASME functions that theoretically spans the space of functional modeling as it is currently known. FIGURE 1: A FUNCTIONAL BASIS [8] As shown in Figure 1, functions have been divided by degree of specification and categorized among eight primary functions. With the assumption that these 42 functions do span the space of functionality of all devices and processes, it follows that all energy-related devices and processes could be categorized by noting which of these functions they accomplish. It is important to observe, however, that all functions listed in the functional basis cannot effectively be applied to energy. For example, the functions listed under the primary function “support” do not easily map into the energy domain. “Stabilize,” “secure,” and “position” are all functions which seem to refer to physical objects. Since energy itself does not have mass, nor is it a physical object, the idea of trying to “secure” energy does not make sense. Additionally, there are primary functions that could apply to energy, but whose secondary and tertiary functions cannot. “Channeling” energy, for example, does make sense, as in channeling electricity through a wire. “Rotate,” a tertiary expression of “channel,” however, is another example of a function that maps much more easily onto physical objects. The fact that all functions cannot be applied to energy implies that there may exist a smaller subset of functions that span the energy domain. Analysis of Existing Function Structures To discover which of the functions listed in the functional basis are applied most frequently to energy and to validate or refute the six core functions identified through observation, 25 existing function structures were analyzed. The incidence of all functions pertaining to energy was quantified. The function structures used in this analysis were taken from two main sources—a textbook on product design [2] and an online design repository maintained by Dr. Robert Stone and his students at the Missouri University of Science and Technology [11]. The function structures selected for consideration in this analysis use energy and are of a somewhat uniform degree of complexity. An example of a function structure and brief explanation of its meaning follows. FIGURE 2: FUNCTION STRUCTURE OF VIBRATING RAZOR [11] Function structures map three basic types of flows through a product’s usage cycle. The three flows are materials, energy and signals. Figure 2 shows an example of a function structure for a vibrating razor. The two material flows shown are the skin surface that is to be shaved (“Import Human Material”) and a hand (labeled) which holds the razor, and these flows are indicated by bold lines and arrows connecting the blocks that interact with these flows. The two types of energy shown are electrical energy (“Import EE”) and human energy (“Import HE”), and the blocks that interact with these energy flows are connected with lines and arrows (not bold). The electrical energy is stored in a battery, and when the razor is actuated by a button, the electrical energy is converted into mechanical energy. In this case, the mechanical energy is vibration, and the converter is an electric motor with an unbalanced mass attached to it. Then, the mechanical energy is transferred to the unshaven Copyright © 2009 by ASME Downloaded 25 Aug 2012 to 128.83.63.20. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm 5 Copyright © 2009 by ASME surface and aids in separating the human material (separating hair from skin). Human energy gripping the razor is converted to mechanical energy by lever action (the shaft of the razor serves as the lever), applying mechanical pressure to the unshaven skin surface. This mechanical pressure energy also aids in the separation of human material. Note that more than one incidence of the “Convert” function, for example, can be found in this single function structure. The first component of the analysis is the simple tallying of all energy-related functions that occurred within each of the 25 function structures. Figure 3 shows these results. FIGURE 3: INCIDENCE OF ENERGY-RELATED FUNCTIONS The three most frequently occurring energy-related functions for the function structures studied were “Import,” “Convert,” and “Transmit.” This nicely validates three of the six functions identified earlier. In order to identify additional core energy functions, some additional analysis was performed. In Figure 4, functions are grouped into the “primary” functions as listed in the functional basis (see Figure 1 for listing of “primary” functions). Incidence of functions with greater specificity than these eight primary functions is lumped together. FIGURE 4: INCIDENCE BY PRIMARY FUNCTIONS As “Channel” includes Import, Export, and Transmit, it is no surprise that its incidence is the greatest. Also, “Convert” occurs very frequently, as was discussed earlier. What is most notable about Figure 4, however, is the high incidence of functions that “control magnitude” as well as “branch” energy. This high incidence shows that regulation, distribution, and control of energy are important parts of using energy effectively. By lumping regulation, distribution, and control into the term “regulate,” another previously-identified core function is validated. One other piece of interesting information can be parsed from Figure 4 with some added analysis. Only 4 of the 25 function structures represented portable systems. As transportation implies portability, these four items should share some key similarities with the transportation vehicles we hope to study. The “provision” primary function includes the secondary functions “store,” “provide,” and “supply.” Each of the four portable items had some capability to “store” energy. From a butane lighter with its stored energy in the form of liquid butane to the batteries contained within an electric toothbrush, portable items and transportation vehicles alike nearly always have the ability to store energy. For this reason, “store” is one additional function that appears to be very important when discussing devices in the transportation sector. USING THE CORE FUNCTIONS IN DESIGN Now that a reasonable set of core functions relating to energy has been identified, it is possible to use these core functions to describe and catalog existing energy systems. Although it is conceded that the six core functions do not necessarily form a basis set, it appears probable that the vast majority of energy systems currently in existence can be broken into constituent parts that perform these six functions. In addition to the specific components that perform the core functions, energy systems are differentiated by the types of energy (or energy domains) with which or on which they operate. Thus, each component in an energy system can be categorized by the core energy functions it performs and the types of energy with which it interacts. The next important step towards satisfying the research objective is the presentation and communication of this information in an example list. In the following sections, a representation of this information in the form of a color-coded morphological matrix is presented and its usefulness is demonstrated in a design trial. A Morphological Matrix The Energy Morph Matrix (EMM) represents an effort to use the CFM strategy as a tool for engineering design. An energy-focused morphological matrix is developed in an attempt to encompass (in a general way) the total set of basic energy system configurations by categorizing energy system components and underlying physical phenomena based on three (rather than the six previously identified) core functions and six types of energy. The six types of energy selected for inclusion in Copyright © 2009 by ASME Downloaded 25 Aug 2012 to 128.83.63.20. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm 6 Copyright © 2009 by ASME the morph matrix are: radiant, solid mechanical, fluid mechanical, chemical, electrical, and heat energy. Nuclear energy is not included, as it is only employed in a small portion of existing transportation vehicles. Using this morph matrix, different combinations of energy converters, storage media, and transmission channels yield a large number of different energy systems. Storage, Transmission, Conversion – 3 functions Three functions out of the six core functions mentioned previously were selected for headings in the morph matrix. In the functional basis, “import” and “export” both fall under the primary function of “channel” along with “transmit.” For this reason, the more general term “Transmission Channels” is selected to cover all ways by which energy can be moved. It is understood that energy being imported or exported will necessarily proceed through at least one of the listed transmission channels. The only other function omitted from the six core functions previously identified is “regulate.” Although energy regulation is an extremely important part of energy use and an integral component in the CFM strategy, and even though effective energy regulation can result in significant energy savings and gains in efficiency, it is not included in the EMM. In this initial attempt to present a list based on the CFM strategy and to use it as a design tool, a number of simplifications were made so that designers would not be overwhelmed by the information presented. Also, since “regulation” of energy occurs within a single energy domain (unlike “conversion” which changes energy domains), it is assumed that regulation and control of energy flow can be optimized at a later phase in the design process—after energy storage media, converters, and transmission channels have already been selected. The EMM is shown in Figure 5: FIGURE 5: ENERGY MORPH MATRIX Other Simplifications and Omissions The EMM presented in Figure 5 is simplified for use in concept generation sessions. One simplification is the omission of certain elements because they are relatively obscure, and because a list of every single storage medium, transmission channel, and converter would be very long and potentially overwhelming to designers in a concept generation setting. Another key simplification is the omission of multiple output energy domains in the “to” column. Most converters do not convert energy from one energy domain to exactly one new domain. Nearly all, in fact, output some quantity of thermal energy because conversion efficiency is not perfect and this imperfect conversion results in the generation of heat. Again, for the purposes of simplification, only the energy domain to which the majority of energy is converted is listed in the “to” column of the EMM. General Instructions Using the EMM as it is shown in Figure 5 involves identifying energy domains with different colors. Energy exists in six different forms (for the purposes of this analysis) and in order to change from one form to another, a converter is needed. Converters have one input energy domain (shown by color in the “CONVERTERS from” column) and one output energy domain (represented by a colored block in the “to” column). Converters receive energy in the input domain and change it into energy in the output domain. Energy can only move from place to place through the listed TRANSMISSION CHANNELS. These channels are energy domain-dependent as Copyright © 2009 by ASME Downloaded 25 Aug 2012 to 128.83.63.20. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm 7 Copyright © 2009 by ASME indicated by the block color. Energy STORAGE MEDIA are also domain-dependent and color-coded by energy type. The simplest way to understand how the EMM relates to real-world energy systems is through an example. This example will show how to use the EMM to represent a hydrogen fuel cell-powered MAV propulsion system as a chain of elements taken from the EMM. Refer to Figures 5 and 6: FIGURE 6: HYDROGEN FUEL CELL MAV PROPULSION SYSTEM AS A CHAIN OF ELEMENTS TAKEN FROM EMM First, energy is stored in the form of hydrogen. Since hydrogen is in the chemical energy domain, only transportation channels in the chemical domain or converters that take chemical energy as their input domain can appear next in the chain. The hydrogen is transported by physically moving it to a fuel cell (moving chemicals), which is a converter (remaining in the chemical energy domain, represented in the EMM by green color). This movement of chemical energy is shown in Figure 6 by a green arrow. The fuel cell converts the chemical energy into electrical energy and thermal energy. The thermal energy is transferred to the surrounding air by convection while the electrical energy is transported by conduction to a battery, which is a storage medium in the electrical domain. The transmission of each form of energy appears as a red arrow and blue arrow, respectively, in Figure 6. Next, an electric motor converts the electrical energy into rotational mechanical energy which is transmitted to a propeller through mechanical coupling where it is converted into fluid mechanical energy. Strung together into a chain and omitting the transport steps and waste heat, the total system becomes: Hydrogen Fuel Cell Battery Electric Motor Propeller. Use of the EMM in an Idea Generating Session To evaluate the efficacy of the EMM as a design aid, printed copies and brief instructions were given to a group of four graduate student designers in a concept generation session. Another group (a control group) of four students was not given the EMM. The design problem given to both groups was to “develop ways to refuel or re-energize an MAV during a mission (i.e., when the MAV is some distance from the launch point and potentially airborne).” Student designers performed a type of rotational drawing exercise—drawing and annotating potential solutions to the design problem on large pieces of paper. Both groups had the same number of participants, were given the same problem statement and instructions for the rotational drawing exercise, had the same amount of time to complete the design task and look over the materials, and all participants were graduate students in mechanical engineering with research interests in design. Counting Ideas The method used to count concepts generated during this experiment is adapted from an existing method [12]. In the counting scheme employed here, a “concept” is a set of drawings and words which describe a device or set of devices that accomplish one or all customer needs stated in the problem statement. A “feature” is something about the concept that is distinct, contained in the expressed intent of the concept, and derived directly from the concept drawing or text (i.e. no implied features are included). An “idea” is like a “feature,” but although features can have overlapping qualities, ideas are unique and do not overlap. The figure below shows a representative concept drawing. FIGURE 7: REPRESENTATIVE DRAWING FROM CONCEPT GENERATION EXPERIMENT The text and pictures in blue ink shown in Figure 7 above comprise concept 13. To clarify, the text contained in concept 13 is: “launch power packs,” “parachute or balloon,” “power pack,” and “plane picks up packages.” The corresponding features contained in concept 13 are “Fire Power Packs,” “Separate Craft Carrying Energy,” “Use Buoyancy” (for the reference to balloons), “Separate Craft,” “Fire Something From Launch Point,” “Use Parachute to Slow Descent,” “General Stored Energy Pack,” and “Mid-air Docking/Fusing/Dividing” (because if the power pack were suspended by a balloon, it would be picked up by the plane mid-air). Clearly, some of these features overlap. “Fire Power Packs” is a more specific version of “Fire Something from Launch Point.” Also, “Separate Craft Carrying Energy” is a more specific version of “Separate Craft.” In order to avoid counting features more than once because of these varying degrees of specificity, so-called “ideas” are defined. For instance, “Fire Something from Launch Point” is the idea which covers “Fire Power Packs,” “Fire MAV,” and “Fire Separate Craft.” Any Copyright © 2009 by ASME Downloaded 25 Aug 2012 to 128.83.63.20. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm 8 Copyright © 2009 by ASME feature which does not overlap with another feature is considered an idea. Results The group that was given the EMM generated a greater number of ideas than the control group, which was not given the EMM. FIGURE 8: NUMBER OF IDEAS PRODUCED BY TWO GROUPS Figure 8 shows the total number of ideas generated by each group as well as each group’s “unique” ideas. Unique ideas were pictured or described by one group only. In total, the EMM group (the one that did receive the EMM prior to concept generation) developed 32 ideas compared to the Control group’s 22 ideas. Of these 32, 19 of the ideas were unique (59%). Of the Control group’s 22 ideas, only 9 (41%) were unique ideas. Therefore, the EMM group not only developed a greater quantity of ideas, but had a greater proportion of ideas that were not developed by the other group. According to this method of counting ideas in this single trial, the EMM showed indications of being a very useful design aid. IMPLICATIONS FOR DESIGN AUTOMATION The structure inherent in the CFM strategy allows for the automatic generation of chains of components and processes which comprise potential systems. Evaluating these potential systems on key performance specifications allows for them to be ranked. There are a number of important metrics for transportation vehicle energy systems including cost, reliability, maintenance requirements, and efficiency which could all be used to rank prospective designs. However, as an illustrative example of a key performance metric in air vehicles, energy density by weight will be used. This metric is important for air vehicles, for example, because typically the more energy that can be stored on the vehicle per added weight, the greater the vehicle’s endurance will be. Given that a large number of energy systems can be expressed simply as chains of storage media and converters, a method to identify and then rank these chains based on key performance specifications or metrics is useful. By identifying and then ranking potential systems, the best performing systems can be selected for further analysis or prototyping, and the large number of non-viable systems can be eliminated. What follows is a plan for how the EMM could be used in combination with a search algorithm and a ranking metric to automatically find energy systems with high energy density. Chain Identification by Search Algorithm A tree-search algorithm could be written that would start with a given storage medium (in a particular energy domain) and then move among a finite number of converters to arrive at some destination energy domain. For example (refer to Figure 5), starting in the thermal energy domain, the converters that could be used are a general endothermic reaction, a thermoelectric generator, a shape-memory alloy, or a Stirling engine. These four converters would be considered as four branches in the tree-search algorithm. Then, from the energy domains to which these converters convert, new sets of branches would appear. The algorithm would store each converter in a chain and then record each subsequent step, generating a large number of possible chains of converters. By capitalizing on the structure inherent in the EMM, a computer algorithm could output a very large number of viable energy systems (chains of storage media and converters). Chain Ranking by Total Energy Density Metric Once chains are identified by the search algorithm, assuming that key performance parameters for each component in the chain are known, a metric can be evaluated for each chain. Consider again the example of the hydrogen fuel cell MAV propulsion system (Figure 6). Two parameter values would first need to be chosen in order to develop an energy density metric for such a system: total energy output over the entire mission and maximum power output. The total energy output (at the propeller) would be a smaller quantity than the total energy stored in the storage medium (the hydrogen) due to conversion efficiency losses in the fuel cell, motor, and propeller. It is necessary to choose a value for the total energy output so that the appropriate quantity of energy storage can be determined. Then, the energy density of the system (by weight) is the total energy stored in the storage medium divided by the total weight of the system. The second parameter value, maximum power output, is needed to determine the total weight of the system. Converters must be scaled properly in order to convert sufficient power to achieve the maximum power output. Conversion power density values (converted power output divided by converter weight) for each converter as well as the maximum power output for the system determine the sizes that the converters must be. The total weight of the system is comprised of the weight of the storage medium and the weights of the converters. By linking energy storage density, conversion power density, and conversion efficiency data to the storage media and Copyright © 2009 by ASME Downloaded 25 Aug 2012 to 128.83.63.20. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm 9 Copyright © 2009 by ASME converters listed on the EMM, and choosing target values for total energy output and maximum power output of the system, a system energy density by weight could be automatically calculated for every viable combination of storers and converters, using a computer tool. CLOSURE The Core-Function Modeling strategy is developed as a way to organize the vast design space of energy systems into a relatively small number of categories (core functions) and energy types. By categorizing an energy system component by the core function it performs, its importance within a given energy system can be better understood. Lists and databases based on this organizational framework can be very useful in energy system design; a variety of components or combinations of components can often be used to accomplish any particular function, and lists of these components can help reduce the amount of searching that a designer must do to find embodiments that perform the particular function. In a single experimental trial, a group that received the Energy Morph Matrix prior to performing a rotational drawing concept generation exercise produced a greater number of ideas than a control group which did not receive the EMM but was charged with solving the same design problem. This result is very promising and future experimental trials should provide more definitive proof that the EMM is a useful concept generation aid. Also, utilizing an expanded EMM that includes more obscure elements or cutting-edge experimental technology might yield interesting results in future experimental trials. The Core-Function Modeling strategy also has implications for design automation because by imposing a simple structure onto the previously amorphous design space of energy systems, a path to using computer tools to aid in the design of energy systems is cleared. Simple search algorithms can move through a list of energy system components and processes (such as the EMM) and unearth viable configurations. The number of configurations can be reduced by ranking the various configurations according to their ability to meet design requirements (such as efficiency, weight, cost etc.). By reducing the need for humans to spend time performing this search, a greater portion of available human resources can be employed in more complex design tasks. Another advantage of the CFM structure is its ability to incorporate new developments in technology. A modified EMM which incorporates various performance parameters of the components listed could easily and quickly be updated with new performance data or new components. In this way, new technology or new achievements in performance can rapidly be incorporated into theoretical systems (chains of components). ACKNOWLEDGMENTSThis work is partially supported by a grant from the AirForce Research Labs (AFRL/RW, Eglin, FL) and, in part, by theUniversity of Texas at Austin Cockrell School of Engineeringand the Cullen Trust Endowed Professorship in Engineering No.1. In addition, we acknowledge the support of the Departmentof Engineering Mechanics at the U.S. Air Force Academy aswell as the financial support of the Dean’s Assessment FundingProgram. Any opinions, findings, or recommendations are thoseof the authors and do not necessarily reflect the views of thesponsors. REFERENCES[1] Hubka, V., Andreasen, M., Eder, W. and Hills, P. (advisoryeditor), 1988, Practical Studies in Systematic Design,Butterworths, London. [2] Otto, K. and Wood, K., 2001, Product Design: Techniquesin Reverse Engineering and New Product Development,Prentice Hall, Upper Saddle River, NJ. [3] Pahl, G. and Beitz W., 1984, Engineering Design: ASystematic Approach, Design Council, London. [4] Ullman, D., 1997, The Mechanical Design Process,McGraw-Hill, New York. [5] Ulrich, K. T., and Eppinger, S., 1995, Product Design andDevelopment, McGraw-Hill, NY. [6] Zwicky, F., 1969, Discovery, Invention, Research Throughthe Morphological Approach, The Macmillian Company,Toronto. [7] Ritchey, T., 1998, “General Morphological Analysis Ageneral method for non-quantified modeling,” 16th EUROConference on Operational Analysis, Brussels. [8] Hirtz, J., Stone, R. B., McAdams, D., Szykman, S., andWood, K. L., 2002, "A Functional Basis for EngineeringDesign: Reconciling and Evolving Previous Efforts," Journal of Research in Engineering Design, 13, (2), pp.65-82. [9] Stone, R. B. and Wood, K. L., 2000, "Development of aFunctional Basis for Design," ASME Journal of Mechanical Design, 122, (4), pp. 359-370. [10] Kurfman, M., Stock, M. E., Stone, R. B. Rajan, J., andWood, K. L., 2003, “Experimental Studies Assessing theRepeatability of a Functional Modeling Derivation Method,” ASME Journal of Mechanical Design, 125, (4),pp. 682-693. [11] MissouriS&T;DesignRepositoryhttp://function2.device.mst.edu:8080/view/index.jsp [12] Linsey, J. S., Green, M. G., Murphy, J. T., Wood, K. L.,Markman, A. B., 2005, "Collaborating to success: Anexperimental study of group idea generation techniques,” 17th International Conference on Design Theory and Methodology, Power Transmission and Gearing Conference, pp. 277-290 Copyright © 2009 by ASME Downloaded 25 Aug 2012 to 128.83.63.20. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm