Document
Evaluation of Electrical System
Requirements for Implementing Turbine
Electrif i ed Energy Management
1 2 3
Jonathan L. Kratz , Dennis E. Culley , and George L. Thomas
NASA Glenn Research Center, Cleveland, Ohio, 44135 Turbine Electrified Energy Management (T EEM) is a concept concerned with the management of energy in an electrified propulsion system . The management of energy in the hybrid - electric architecture has potential to benefit the turbomachinery and the aircraft it powers . The concept is particularly useful for improving operability during transient operation and could be leveraged to design a better performing engine . The concept utilizes electric machines coupled to the engine shafts and an electric power distribution sy stem that includes energy stor age . A controller is used to decide when and how energy is moved around the electrified propulsion system, particularly when considering energy conversion between mechanical and electrical forms. Prior work has shown that the electric machines can be used to supply / or extract supplemental power to /from the engine shafts to improve their operability and achieve or enable propulsion efficiency and performance benefits. However, t he previous studies did not consider the practical constraints of the electric al machines and energy storage devices that are required for implementing the TEEM system architecture concept . This paper presents an integrated engine a nd electrical system model that is used to evaluate the electrical system requirements . The model capt ure s the physics of the conceptual , Advanced Geared Turbofan 30,000lb f (AGTF30) engine , which features advanced technologies such as a compact gas turbine and a variable area fan nozzle . For this work, t he engine is aug me nted with electrical system components that allow for the implementation of the TEEM concept .
The evaluation presented suggests the potential of the TEEM concept to provid e performance benefits for a turbofan engine .
Nomenclature Alt = altitude , ft C = bank capacitance , F C = capacitance of an individual super - capacitor cell or module , F SC E = energy capacitor , kW - hr E = useable energy capacity , kW - hr U E = energy capacity of an individual super - capacitor cell or module , kW - hr SC E = total energy capacity , kW - hr T I = maximum continuous current of an individual super - capacitor cell or module , A Cont,SC I = maximum peak current of an individual super - capacitor cell or module , A Peak,SC N = shaft speed , rpm N = high pressure spool speed , rpm HPS N = low pressure spool speed , rpm LPS P = power , hp SF = super - capacitor energy capacity safety factor E Research Engineer, 21000 Brookpark Rd., Cleveland OH, MS - 77 - 1, AIAA Member .
Research Engineer, 21000 Brookpark Rd., Cleveland OH, MS - 77 - 1, Senior AIAA Member .
Research E ngineer, 21000 Brookpark Rd., Cleveland OH, MS - 301 - 5, AIAA Member.
SF = super - capacitor current safety factor I SF = super - capacitor voltage safety factor V SM = stall margin , % SM = baseline stall margin , % b SM = stall margin benefit expressed as a percentage of the stall margin improvement enabled by the electric ben m achines , % SM = stall margin with unconstrained electric machines , % u τ = torque , ft - lb f τ = torque applied by the electric machine on the high pressure spool , ft - lb HPS f τ = torque applied by the electric machine on the low pressure spool , ft - lb LPS f V = voltage of the super - capacitor bank when it is considered to be fully charged , V FC V = absolute maximum bank voltage , V Max V = maximum voltage of an individual super - capacitor cell or module , V Max,SC V = minimum bank voltage , V Min I. Introduction ASA is investigating the electrification of propulsion systems for commercial air transports, a sector that has previously been dominated by gas - driven turbomachinery. NASA has developed various hybrid - electric aircraft
N
concepts that integrate turbomachinery with electrical hardware. These concepts often involve the use of electric mac hines (EMs) , coupled to one or more of the engine shafts , to extract or supply power. Several of the concepts include energy storage , such as batteries , and consider a vari e ty of electrical system architectures. Some concepts involve the use of electricall y driven propulsors located in strategic positions on the aircraft to achieve aerodynamic benefits and higher propulsive efficiency . Ultimately, the conceptual studies have shown benefits with respect to reduced fuel burn, lower emissions , and/or reduced n oise [ 1 - 5 ]. While the electrification of aircraft propulsion systems can produce benefits for the aircraft, it can also enable benefits for the turbomachinery that inevitably fold into aircraft level benefits such as reduced thrust specific fuel con sumption . Turbine Electrified Energy Management (TEEM) is a concept that seeks to do just that by managing how energy is supplied to or extracted from the engine shafts based on the state of the engine and the demand from the pilot. . While TEEM was motivat ed by the emergence of hybrid electric propulsion concepts that already possess much of the electrical components necessary for implementation, the authors believe this technology to be applicable to more traditional gas turbine engines in the near term. T his would demand that hardware necessary for implementing TEEM , economically justify its addition to the engine architecture if it is not already present. Such hardware may include electric machines (motor s/generators), energy storage devices (ESDs) such a s batteries and super - capacitors , and supporting electrical devices such as inverters , rectifiers , direct current – direct current (DC - DC) converters, cables, resistors, etc. It may also encourage dual use of existing components such as an electric starter m otor , ESDs , and auxiliary power units (APUs) .
At it s broadest leve l, TEEM addresses the management of energy in an electrified turbine - based propulsion system.
EMs coupled to the shaft(s) of the turbomachinery are used to selectively supply or extract to rque in order to actively al ter the operation of the engine. Thus, the TEEM concept can be viewed as an oper ability technology. Potential benefits of power addition/extraction during steady - state operation include , but are not limited to , the reduc ed need for stability bleeds and variable geometries, and exp ansion of the operating range of the engine . Power supplied/extracted during transients can alter the dynamic response. Done properly, transient stability margin can be reduced or even eliminated , thus e nabling more efficient and better performing engine designs.
In prior work [ 6 ] , the feasibility of the TEEM concept was demonstrated through application to the NASA - developed parallel hybrid - electric turbofan (hFan) [ 7 , 8 ] concept. While this was a modeli ng and simulation effort, it illustrated the anticipated impact of the technology. The documented study illustrated the elimination of the low press ure compressor stability bleed , or variable bleed valve (VBV) , through energy management on the shafts . It a lso illustrated the application of EMs to selectively supply or extract power during engine transients in order to improve transient operability. A reduction or complete elimination of undershoot in compressor stall margin (SM) during engine transients was observed . Whereas open - loop torques were applied by the EMs in Ref. [ 6 ] , Ref. [ 9 ] describes a closed - loop control approach for applying TEEM to a conceptual gas tu rbine engine model. The engine is known as the Advanced Gear ed Turbofan 30,000lb (AGTF30) [ 10 ], a two - spool gear ed turbofan engine representative f of technology expected to enter service in the 2030 - 2035 time frame. The engine features a compact gas turbine and ® a variable area fan nozzle (VAFN) . The dynamic engine m odel was executed in the MATLAB/Simulink environment. It was developed using the Toolbox for Modeling and Analysis of Thermodynamics Systems (T - MATS) [ 11 ] and was built to match the performance of the NASA N+3 reference engine [ 12 ] implemented in the Numeri cal Propulsion System Simulation (NPSS) code [ 13 ] . This paper exp an ds on the work documented in Ref. [ 9 ].
The emphasis of Ref. [ 9 ] was the control approach for implementing aspects of the TEEM con cept, but it made some assumptions about the electrical system , without consideration of hardware requirements , that may have resulted in an over - designed or under - designed system . Similarly it did not consider the limitations of the hardware that could re sult in unrealizable performance. This paper continues with the same application and control approach es discussed in Ref. [ 9 ], but it focuses on identifying basic requirements for the electrical system in order to assess the p racticality of applying the TEEM concept on the chosen propulsion system . Requirements were developed through parametric studies that ultimately prompt a number of recommendations for the electrical components that either directly or indire ctly imply sever al constraints . The study is focused on the requirements for the EM(s) that affect the shafts and the ESDs that supply or absorb the energy to do so .
Ref. [ 9 ] considers 3 configurations of the electrical system that were cha racterized by the number of EMs and variations in the shaft (s) the y were coupled to. It considered EMs on both shafts or a single EM on either the low pressure spool (LPS ) or the high pressure spool (HPS). These configuration s are referred to as the dual - spool configuration, LPS configuration, and HPS configuration , respectively. Results from Ref. [ 9 ] showed that the LPS configuration could not practically a ffect the operability of the high pressure compressor ( HPC ) during ac celeration transients. Furthermore, there were circumstances where the configuration required a large amount of power to be dissipated, thus requiring a large bleed resistor bank. For these reasons, the LPS configuration was not considered in this study. H owever, the dual - spool and HPS configurations were evaluated to assess if either configuration imposes different requirements. Two varia nts of the dual - spool configuration w ere considered. The first varia nt was consistent with the implementation in Ref. [ 9 ]. The second vari ant differed in the way it handled excess power dissipation during deceleration transient s . The first varia n t sen t the excess power extracted from the LPS to the ESDs until the ESDs were near full charge . Any additional power would be applied by the HPS EM. The second variant does not charge the ESDs during decelerations transients, but simply applies the excess power to the HPS spool via the HPS EM. It was hypothesized that this approach would reduce the powe r requirement of the LPS EM , thus reducing its size . To distinguish the two variations, t hey will be referred to as dual - spool configur ation vari a n t 1 (v1) and dual - spool configuration varia nt 2 (v2). In all configuration s and varia nts , the EMs are assumed to spin at the same speed as the shaft they are coupled to.
A presentation of this paper begins with Section II. In Section II of this paper, a description of the models used in the study is presented . I nclude d in the description is a discussion of the electrical system, its modeling, and its integration with the engine model. Section III provides a brief description of the TEEM control strategy. Section I V describe s the studies conducted with the model and the results from those studies. Section V discusses tradeoffs regarding the TEEM system configurations and potential dual uses of TEEM hardware . Finally, Sections V I provides some concluding remarks.
II. System Modeling and Integration In this study, two different model s a re utilized. Both apply the control strategy discussed in Ref. [ 9 ]. The controller used to apply this strategy will be referred to as the TEEM controller. The first model captures the physics of the AGTF30, but it lacks fid elity with respect to the electrical system . The model includes handles for affecting torque on the engine shafts , but does not model how the electrical system provides the required torque . It simply assume s that the electrical system supplies the torque t hat the engine needs. This model will be referred to as the engine - centric model. The second model is the engine - centric model integrated with electrical system component models. It will be referred to as the integrated model.
A. Engine - Centric Model The AGTF 30 engine represents a two - spool geared turbofan with a compact gas turbine and a VAFN . The engine model has been modified in a number of ways. Most notably it has been modified to accept inputs for supplemental torque on th e engine shafts. First order dyn amics, saturation limits, and rate limits were applied to the torque co mmanded by the TEEM controller to simulate EM dynamics . In all case s the EMs w ere assumed to have a bandwidth of 35 rad/sec. While much lower bandwidths were initially tested, the range of practical values did not have a significant impact on the ability of the TEEM controller to a ffect operability. The EM torque is supplied directly to the engine as an amount that is either added to or extracted from the shaft. The model does not consid er how the EMs are supplied power, or how excess power extracted by the EMs is used or dissipated . The model simply assumes that the require d power is supplied and that all excess power is absorbed or dissipated. T he main parameters were the EM torque satu ration limits and rate limits. The model was used to identify basic size and responsiveness requirements for the EMs while also implying requirement s for the rest of the electrical system. Since the responsiveness of the engine degrades over time, it is im portant to consider how this may impact the needs of the electrical system over the lifespan of the engine. Therefore, the engine - centric model incorporates a deterioration model that permits the adjustment of component health parameters in order to accoun t for degradation . The deteri or ation model was adapted from the model applied to the Commercial Modular Aero - Propulsion System Simulation 40,000lb (C - MAPSS40k) engine model f [ 14 ]. In the analysis to follow , a worst - case fully deteriorated engine was assumed . Degradation of electrical components was not considered in this study.
B. Integrated Engine and Electrical System Model The critical components of the electric al system are the EMs and ESDs. However, there will also be various component s to support the t ransfer of power between these devices . While these components were not the focal point of this study, they were modeled with the primary purpose of introduc ing realistic efficiencies/losses in to the power transmission . These components can be grouped into a few categories. The first category is motor controller element s , which include s functions such as variable frequency inverters (DC - AC conversion) and rec t ification of alternating - current (AC - DC conversion) . The second category is converters . This catego ry may include devices such as direct current to direct current (DC - DC) converter s and transformers. The third category is energy management elements and it may include devices such as bleed resistors and current /voltage limiting control circuits. The four th and final category is power transmission elements which primarily consists of cables. A high level schematic of the electrical system is shown in Fig. 1 , which assumes a DC power distribution system that includes a DC - DC c onverter .
A particular electrical system architecture is assumed in this study, however it should be noted that the architecture and component parameters are merely a starting point for the systems design. Recall that the purpose of these studies is to eva luate feasibility through identification of electrical system requirement s , not to perform a detailed system design. In a real world application, the implementation should be reconsidered to find the optimal system design . The architecture in this work con sists of one EM for the HPS configuration and two EMs for the dual - spool configuration variants. The EMs can function as either a motor or generator. Energy storage is provided solely by a bank of super - capacitors . The power is transmitted through a DC bus whose voltage is regulated by a DC - DC converter. The EM control unit converts power from DC to AC to drive the EM(s) as motor(s) and from AC to DC when the EM(s) are acting as generator(s) . Various cables connect the devices and transmit power through the ir conductors as needed.
The electrical system was modeled using the Electrical Modeling and Thermal Analysis Toolbox (EMTAT).
EMTAT is currently under development at the NASA Glenn Research Center and has evolved from the electrical system modeling ef fort documented in Ref. [ 15 ] to perform power flow analysis on hybrid - electric systems. The toolbox is a Simulink library of electrical system components for capturing the interactions of turbomachinery and electrical systems at an appropriate level of fide lity. Such interaction s include : (1) the coupling of EM torque/power dynamics with turbomachinery shaft dynamics , (2) charge and discharge cycles of ESDs , and (3) overall dynamic flu ct uations of currents and voltages in the electrical system that occur in response to changes in electrical loads. The modeling time scale is often on the order of milliseconds, similar to the time scale used to model the turbomachinery Fig. 1 TEEM electrical system with the HPS EM acting as a motor and the LPS EM acting as a generator [ 9 ] Fig. 2 Integrated engine and electrical system m odel dynamics. Faster phenomena , such as those Table 1 . Basic parameters of the electrical system model associated with switching circuits within the Component Parameter Value power system , are not modeled . However, Super - Capacitor Capacitance -- bulk efficiencies are applied as an effort to Bank Full Charge Voltage -- approximate their overall impact .
Equivalent Series 0.0031 Ω The EMs are modeled with a const ant Resistance impedance and a motor map that defines its Leakage Current 0.04 A efficiency at different speeds and torques.
DC - DC Converter Bus Voltage 1000 V The inverter a nd rectifier models produce an Maximum Efficiency 0.98 output voltage that scales with frequency Bleed Resistor according to the specified flux const ant. The Resistance 0.5 Ω Bank load is assumed to have no inductance , Power Supply thereby implying a unity power factor. The Resistance 0.001 Ω Cable efficiency of the inverter /rectifier is specified Motor Cables Resistance 0.001 Ω by a look - up table that maps output current LPS EM Maximum Power -- and voltage to efficiency. The DC - DC Maximum Torque -- converter model is meant to produce a Ma ximum Torque Rate -- const ant output voltage. Its efficiency is Maximum Speed 7500 rpm determined by a look - up table that is a Maximum Efficiency 0.96 function of input voltage and current. The input voltage is calculated such that power is Resistive Impedance 0.01 Ω conserved through the element . The super - HPS EM Maximum Power -- capacitor bank is modeled with a specified Maximum Torque -- capacitance, equivalent series resistance, and Maximum Torque Rate -- a constant leakage current. The rate of change Maximum Speed 23000 rpm in voltage of the capacitor is computed Maximum Efficiency 0.96 numeric ally inside the Simulink block given Resistive Impedance 0.01 Ω the voltage from the previous time - step . The L PS EM Inverter Voltage/Flux Constant 0.996 V/Hz v arious cables in the system are modeled and Rectifier Maximum Efficiency 0.98 with a simple impedance or resistance. Each HPS EM Inverter Voltage/Flux Constant 0.307 V/Hz block that represents an electrical component and Rectifier Maximum Efficiency 0.98 has an associated error and an independent variable that is typically a current or voltage associated with that component . A Newton - Raphson solver is used to adjust t he independent variables to satisfy error tolerances. Dynamic variables , such as the super - capacitor voltage , are pulled out of the i terative solver s ubsystem to enforce the dynamics.
The overall structure of the electrical system model and its coupling to the engine model are presented in Fig. 2 .
The electrical system model consists of 3 l evels: (1) th e electrical component models, (2) the solver, and (3) the dynamic s layer. The electrical system model receives the shaft speeds and electrical machine torques from the engine model. Prior to entering the electrical system model, first order dynamics, satu ration limits, and rate limits are applied to the commanded torques. Thus the torques input to the electrical system model are the to rques that are applied to the shafts . S ome of the outputs of the electrical system are utilized by the TEEM controller, inc luding the voltage across the super - capacitor bank or its state of charge .
The efficiency of the electrical machines was defined by maps relating the efficiency to the torque and speed.
Similarly, the efficiency of the DC - DC converter and the inverters/re ctifiers were defined by maps relating the efficiency to voltage and current. The maps were produced from an empirical characterization of a model of the Subsonic Ultra Green Aircraft Research (SUGAR) Volt hFan power system as presented in Ref. [ 16 ]. The ba sic parameters of the electrical system model are given in Table 1 . The parameters that do not have a n assigned numeric value were altered during the parametric study . Parameters having a numeric value remai ned fixed through out the analysis.
III. TEEM Control Strategy The TEEM control strategy in Ref. [ 9 ] has 2 primary functions : (1) improving transient operability , and (2) re - charging the ESD s . These goals must consider operability constraints, particularly at low power settings, thus requiring coordination with various actuator control schedules. Shown in Figure 3 is a high level schematic meant to capture the essence of the control approach. Note t hat “ v1 ” in Fig. 3 b refers to dual - spool configuration v1 , while “v2” refers to dual - spool configuration v2 . The white blocks signify high - level decision - making criteria while the blue blocks signify some action that is im plemented if the appropriate criteria are met.
During transients, transient operability is prioritized while charging functions are temporarily suspended , and the low power stability issue is handled solely by the VBV , as it is in the baseline AGTF30 mode l . A c tive control of the EM(s) is applied via a gain - scheduled proportional integral (PI) controller that seeks to match the steady - state shaft speed vs. fuel flow or the low pressure compressor (LPC) pressure ratio (PR) vs. fuel flow relationship. LPC PR control is only applied during deceleration transients when there is a single EM on the HPS. The proportional and integral gains are scheduled bas ed on Mach number (MN), altitud e (Alt) , and power lever angle (PLA) . PLA is commanded by the pilot of the airc raft and corresponds to the engine’s throttle position or power setting . In either configuration, power is input to the HPS during accelerations. For the dual - spool configuration , for which an EM is present on the LPS, power is extracted from the LPS durin g decelerations. Dual - spool configuration v1 directs the Fig. 3 High level schematic summarizing the TEEM control approach where (a) captures the HPS configuration and (b) captures the dual - spool configuration extracted power to the ESDs until they are nearly charged and then begins expending the excess power by applying it to the HPS via the HPS EM. Dual - spool configuration v2 does not charge the ESDs , bu t instead appl ies all of the extracted power to the HPS via the HPS EM . The HPS configuration deals with both acceleration and deceleration transients by applying power to the HPS via the HPS EM .
During steady - state operation, the VBV solely handles the l ow power stability issue wh ile the ESDs are sufficiently charged . When the devices are not sufficiently charged, an EM is used to extract power from one of the shafts to charge them. The amount of power extraction is set by a schedule. The power extraction is the same for most conditions but tapers to zero as the ESDs approach full charge . It is also reduced for very high engine power and very high altitude in order to preserve maximum thrust and ensure operability at high altitude . If an EM is present on t he LPS then use of that device is the preferred means of power extraction. In either case, the VBV and VAFN schedules have been modified in order to maintain the same level of LPC and fan operability as the baseline AGTF30 while power is extracted from eac h shaft . For more details or clarification of the TEEM control strategy, refer to Ref. [ 9 ]. All parametric variants simulated in this study utilize the control logic presented here .
IV. Parametric Stud ies A. Evaluation of EM Require ments The EM requirements were considered first. The goal was to identify the maximum power, torque, and torque r ate required for all applicable EMs , for each configuration considered . For conservative approximations, an end - of - life engine w as assumed in t he analysis . The first step was to identify the stall margin provided by the baseline engine and TEEM implementation via unconstrained EMs. In other words, the minimum stall margin achieved during transients with the baseline engine and with application of the TEEM concept using EMs that can instantaneously supply or extract whatever power is commanded . The difference between these values provides an indication of how much benefit the TEEM concept could potentially provide regarding the transient operabilit y of the engine. When the EM is constrained, its benefit is expressed as a percent benefit as defined by Eq. ( 1 ).
𝑆𝑀 − 𝑆𝑀 𝑏 𝑆𝑀 = × 100% ( 1 ) 𝑏𝑒𝑛 𝑆𝑀 − 𝑆𝑀 𝑢 𝑏 In Eq. ( 1 ), SM is the minimum stall margin when the EM is constrained, SM is the minimum stall margin when the u EM is unconstrained, and SM is the minimum stall margin of the baseline engine. This definition allows the results b at various operating points t o be normalized. In addition it provides a clear measure of how close the constrained results ar e to achieving the full benefit , thereby permitting identification of the EM requirements.
To determine the EM requirements, two set s of simulations were cond ucted, both with the engine - centric model.
First, simulations were conducted for 20 select design points throughout the flight envelop e to produce data that could be used to infer the EM requirements .
Those points are identified in Fig. 4 by red plus signs. Second , simulation s were conducted for 1000 test points throughout the flight envelope to verify that the TEEM controller was able to achieve the expected operability improvements with the sele cted EM constraints . Those tes t points are identified in Fig. 4 by blue x’s . For eac h Fig. 4 AGTF30 flight envelope with design points used to size the design point or test point, simulations EMs and test points used to verify sufficient EM performance were run for an idle to full throttle burst throughout the flight envelope (acceleration) and a full throttle to idle chop (deceleration). In either case, the t hrottle position was ramped from its initial value to its final value over the course of 1 s. The first set of design point simulations varied the constraint on the maximum torque of the EMs . At the same time, constraints were placed on the maximum power setting that the EMs could apply and extract. The maximum HPS EM power constraint , in horsepower (hp) , was set to be 3.75 times the maximum torque constraint in ft - lb . Similarly the maximum LPS EM power constraint , in hp , was set to be 1.25 times the max imum torque constrain t f in ft - lb . The scaling values of 3.75 and 1.25 were selected because they are approximately equal to the expected f speeds of the respective shafts divided by the unit conversion factor of 5252 in Eq. ( 2 ), where P is the power in hp and N is the speed of the shaft in rpm. Thus , they are approximate scaling factors that relate the EMs power and torque.
𝑁 ( 2 ) 𝑃 = 𝜏 A root finding technique, known as the bisection root method, wa s used to numerically adjust and identify the minimum torque required to achieve 98%, 95%, and 2 0% of the stall margin benefit. For accelerations, the HPC SM was used for the SM calculation and the HPS EM torque limit was adjusted by the root finding me thod. During ben decelerations , the LPS SM was used for the SM calculation and the HPS EM or LPS EM torque limit was adjusted ben by the root finding method , dependent up on the configuration. Once the torques were identified, simulations were conducted for a nu mber of torque constraints between the 2 0% and 95% solutions in order to complete the data set.
This data was used to produce the plots shown in Fig. 5 , which w ere used to subjectively choose the minimum torque requirement for each EM. The r ecommendation s for achieving nearly all of the benefit are provided in Table 2 .
The maximum torque s for the EMs were fixed at the values identified previously Simulations were run at the same design points in the same m anner as before, but this time the maximum torque rate constraints were varied instead of the maximum torque constraints. Once again , the bisection root method was used to numerically identify the minimum Fig. 5 Stall margin benefit versus EM torque limits for the HPS configurati on (a), dual - spool configuration v1 (b), and dual - spool configuration v2 (c) Table 2 . Electric machine requirements Electric Machine Maximum Maximum Power , Maximum Time to achieve Torque, ft - lb hp Torque Rate, ft - lb /s Maximum T orque, s f f HPS EM 200 750 250 0.8 LPS EM 550 610 150 3.7 (dual - spool v1) LPS EM 350 410 95 3.7 (dual - spool v2) torque rate required to achieve 98%, 95% and 2 0% o f the stall margin benefit at each of the test points for both the acceleration and deceleration transients. Simulations were run with different torque rate constraints between the 20% and 95% solutions in order to complete the data set. The results from t hese simulations are shown in Fig. 6 . From these plots , the recommendations for maximum torque rate were inferred and recorded in Table 2 . The maximum torque and torque rate allow for the minimum response time of the EM to be inferred. These values are recorded in Table 2 and they represent the approximate minimum time s to achieve the maximum torque output when starting from zero torque . EMs are known to have a very quick torque re sponse time. Response times on the order of seconds, as required here, are not expected to be an issue.
To verify that the constraints identified for the test cases are valid , the constrain ts were applied for the 1000 random flight envelope points indica ted in Fig. 4 . Figure 7 presents plots of the HPC SM and LPC SM improvement over the baseline engine for these simulations. As can be seen, substantial SM benefit s are achieved throug hout the flig ht envelop e . With the enforce d constraints, implementation of the TEEM concept would allow for 2.5 % to 5.75 % HPC SM and up to 3.75 % LPC SM to be removed from the transient operability stack. Note that at high Mach number and low altitude, the LPC SM improv ement approaches zero because the LPC SM undershoot for the baseline engine Fig. 6 Stall margin benefit versus EM torque rate limits for the HPS configuration (a), dual - spool configuration v1 (b), and dual - spool configuration v2 (c) Fig. 7 HPC SM (a) and LPC SM (b) improvements over the baseline engine with constrained EMs approaches zero. Therefore, the potential for improvements approach zero. Furthermore, the steady - state stall margin at these flight conditions includes additional margin in comparison to other flight conditions . While the minimum LPC SM is reduced by < 0.1% in some instances with the HPS configuration and dual - spool configuration v2, this small decrease to operability is not of concern . In addition , these operation al considerations are not typical of a commercial transport mission and t herefore would be unlikely to be encou n tered . The simulation data was analyzed to see if the maximum power constraint was reach ed. If not, the power constraint enforced previously could be relaxed. While the HPS EM power constraint was reached in the simulations, the LPS EM power constraint was not. This prompted a relaxation in the maximum power requirement for both of the dual - s pool variants. The final recommendations on the maximum power requirement are recorded in Table 2 .
Recall that the recommendations for EM requirements were made assuming a worst - case end of life engine. It is worth noting that the requirements would have been substantially lower had the analysis Fig. 8 Stall margin benefit versus EM torque limits been conducted for a new engine. To illustrate this for a new engine at the worst observed design points point, Fig. 8 shows the stall margin benefit versus torque limit for a new engine at the design point at whi ch the most torque was required for the end of life engine . The torque and power limits were reduced to 120 ft - lb f (~450 hp) for the HPS EM in all configuration s. The LPS EM torque and power limit was reduced to 420 ft - lb (~465 f hp) for dual - spool configur ation v1, and 250 ft - lb (~290 hp) for dual - spool configuration v2 . This equates to a f reduction in power requirements of 30 – 40%.
B. Evaluation of Super - Capacitor Requirements Once the EM requirements were determined , focus was directed to ward determining the super - capacitor requirements . The super - capacitors need t o collectively supply or absorb power as the EMs demand . This means that they need to be capable of storing a sufficient amount of energy to address engine transients while also handling large pe ak currents. The super - capacitor bank was assumed to be structured into n strings of super - capacitors that are in p parallel with each other. Each string has n super - capacitor s in series. It is possible that each super - capacitor in the s bank could consist o f multiple cells arranged in a similar fashion. Some consideration was given to redundancy for fault handling. If a string should fail, the remaining strings should be able to handle the current demand and there should be enough energy capacity t o handle a worst - case transient . Also, if a cell within a string were to short circuit, the other cells within th at string should be able to safely increase in voltage such that the desired bank voltage can be achieved. To address these potential situations, curr ent, energy, and voltage safety factors ( SF , SF , and SF ) were I E V applied as defined by Eq. ( 3 ) and ( 4 ) .
𝑆𝐹 = 𝑆𝐹 = 1 + ( 3 ) 𝐼 𝐸 𝑛 − 1 𝑝 𝑆𝐹 = 1 + ( 4 ) 𝑉 𝑛 − 1 𝑠 For this study, t he peak current of the individual super - capacitors , I , was assumed to be 1900 A based on the P eak,SC capabilities of super - capacitors modules that are currently available [ 17 ]. The minimum amount of useable energy that the super - capacitors should have is a bit subjective. For this analysis it was assumed that the super - capacitors should have enough useable energy capac ity to handle a worst - case transient for an end - of - life engine . It is a bit subjective as to how to define the worst - case transient. Considering the 1000 test cases r u n during the analysis of the EM requirements, the maximum energy required for an accelera tion transient was ~1.3 kW - hr. The worst - case point occurs at high altitude and high Mach number , a scenario in which a n extreme acce leration transient is un likely. Thus, there could be an argument to narrow the search for a worst - case scenario down to mor e likely scenarios, such as take - off or an aborted landing. Narrowing the search range for a worst - case scenario to Mach number s below 0.3 and altitudes below 1 5,000 ft, the amount of useful energy Table 3 . Super - capacitor recommended useful energy capacities capacity could be reduced to Worst - case energy used during transients, kW - hr ~0.8 kW - hr. During the Worst - case search range Acceleration Deceleration (HPS Total (HPS decele rations, the HPS configuration configuration configuration applies power, only) only) which could arguably drive up Entire Flight Envelope 1.3 2 3 its energy capacity MN < 0.3, Alt < 10,000 ft 0.8 1.6 2.5 requirements . Considering the entire flight profile, ~2 kW - hr of energy is applied during the worst - case deceleration transient. For the reduced search range, t he energy drops to 1.6 kW - hr. To address a scenario in which the pilot could rapidly reduce the power immediately following an accelerati on transient, the total amount of useful energy capacity should be greater than 3 , and 2.5 kW - hr when considering the entire flight envelop and the reduced search range , respectively. Once again, this would be an unlikely scenario. Therefore, an argument could be made to completely disregard the combination of transients and only focus on the transient scenario(s) for whi ch there is a reasonably significant chan c e of occurring . The results are summarized in Table 3 . Based on observation of the LPC SM response, it is suspected that the HPS EM controller , used for deceleration tran sients in the HPS configuration, command ed more power than is necessary to achieve the operability goals. Thus, it was noted that f urther attention to the design of this controller would reduce the ene rgy usage during decelerations.
The following analysis of the super - c apacitor parameters focuses primarily on a useful energy capacity E of 0.8 U kW - hr . Ho wever , the parametric analysis can be easily performed for other values of useful energy capacity following the same analysis proc edure . The e nergy requirements are belie ved to be conservative, noting that energy usage is expected to be much less during typical operation of the engine . To put this into perspective , Ref. [ 9 ] applied the TEEM concept to a mid - life engine and simulated a flight w ith a duration of nearly 1 hr and 40 min. During that flight, only 0. 34 to 1.7 kW - hr was used in total . The largest expenditure of energy during a single transient was ~0.2 kW - hr with a maximum applied power of less than 200 hp .
In the parametric analysi s , n , n , the bank equivalent capacitance C , and the power extraction for charging P were s p c varied to study their impact on the following: minimum bank voltage V , full charge bank voltage V , absolute Min FC maximum bank voltage V , total bank energy capaci ty E , individual super - capacitor maximum voltage V , Max T Max, SC individual super - capacitor energy capacity E , individual super - capacitor capacitance C , and the individual SC SC super - capacitors maximum continuous current I . The full charge bank voltage is defined in this analysis as the Cont,SC voltage at which the super - capacitor bank is considered to be at full charge. This value will be lower than the absolute maximum bank voltage , which is higher to address potential fault scenarios . The calculation of each of t he parameters listed prior will be discussed next along with a visualization of their parametric variation .
The individual super - capacitor capacitance was computed with Eq. ( 5 ).
𝑛 𝑠 𝐶 = 𝐶 ( 5 ) 𝑆𝐶 𝑛 𝑝 F igure 9 plots the relationship to visualize the individual super - capacitors capacitance in order to achieve the desired bank capacitance.
More super - capacitors in series co rrelates to lower bank capacitance while more super - capacitor string s in parallel correlates to Fig. 9 Individual super - capacitor capacitance more bank capacitance. To keep the current demand from exceeding the limitation of the individual super - capacitors, a minimum bank voltage requirement needs to be imposed . This means that a portion of the energy within the super - capacitors should be kept in reserve . The minimum voltage was calculated using Eq. ( 6 ) .
745 . 7 𝑃 / 0 . 9 𝑚𝑎𝑥 𝑉 = ( 6 ) 𝑀𝑖𝑛 𝑛 𝐼 / 𝑆𝐹 𝑝 𝑃𝑒𝑎𝑘 , 𝑆𝐶 𝐼 Here P is the maximum power draw (750 hp) and the factor of max 0.9 is introduced to account for power losses during power transmission. Figure 10 shows the minimum bank voltage plotted as a function of n As n increases, V decreases. This is due to p . p Mi n the current safety factor, which reduces as the number of strings increases because there are more strings to share the current load i f a single string were to fail.
Fig. 10 Minimum bank voltage A conservative limit on the maximum continuous individual super - capacitor current is given by Eq. ( 7 ).
Fig. 11 Maximum continuous super - capacitor Fig. 12 Time to charge current 745 . 7 𝑃 𝐶 𝐼 = ( 7 ) 𝐶𝑜𝑛𝑡 , 𝑆𝐶 𝑛 𝑉 𝑝 𝑀𝑖𝑛 Since transients are short, power addition or extraction during transients should not drive the I requirement. In Cont,SC the current control strategy, charging the super - c apacitors is the only mode of operation that has the potential to require current flow to or from the super - capacitors for an extended period of time . Thus the charging power is used to compute the maximum continuous current. The estimate is conservative b ecause V will increase as super - capacitors Min charge, thus the limit would only be reached for an instant. Figure 11 plots the variation of I with n and P . To Cont,SC p C provide some context for the difference between the different ch ar ging rates, Fig. 12 shows the approximate time required to reach the full charge bank voltage from the minimum bank voltage. This figure also gives an indication of how closely in time extreme transient s can occur . Ideally, s ubsequent extreme transients will be spaced apart such that the super - capacitor has time to re - charge .
The full charge bank voltage must be sufficient to achieve the desired amount of useful energy capacity. The maximum bank voltage must also account for the voltage safety factor defined in Eq. ( 4 ). The total energy capacity of the super - capacitor bank must allow the maximum absolute bank voltage to be achieved . Equations ( 8 ), ( 9 ), and ( 10 ) were used to calculate the values of each of the respective parameters .
2 1 ( 8 ) √ 𝑉 = ( ( 3 . 6 × 10 ) 𝐸 𝑆𝐹 + 𝐶 𝑉 ) 𝐹𝐶 𝑈 𝐸 𝑀𝑖𝑛 𝐶 2 𝑉 = 𝑆𝐹 𝑉 ( 9 ) 𝑀𝑎𝑥 𝑉 𝐹𝐶 ( 10 ) 𝐸 = 𝐶 𝑉 𝑇 𝑀𝑎𝑥 Figure 13 , 14 , and 15 show the variation of each of these parameters with respect to C , n , and n , assuming E to s p U be 0.8 kW - hr. While specific recommendations are not given, some g eneral conclusions sug gest that the higher the number of super - capacitor cells or modules, the lower the overall maximum voltage and total energy capacity.
Also, p utting capacitor s t rings in parallel is more beneficial than putting the capacitors in series.
With a better unders tanding of the bank requirements, a few more individual super - capacitor parameters can be evaluated ; s pecifically , V and E which were Max,SC SC computed using Eq. ( 11 ) and ( 12 ).
Fig. 13 Full charge bank voltage Fig. 14 Maximum bank voltage Fig. 15 Total bank energy capacity 𝑉 = 𝑉 / 𝑛 ( 11 ) 𝑀𝑎𝑥 , 𝑆𝐶 𝑀𝑎𝑥 𝑠 − 7 2 ( 12 ) 𝐸 = ( 2 . 78 × 10 ) × 𝐶 𝑉 𝑆𝐶 𝑆𝐶 𝑀𝑎𝑥 , 𝑆𝐶 Figure 16 , and 17 show the variation of each of these parameters with respect to C , n , and n assuming E to be 0.8 s p U kW - hr. The maximum voltage of the individual super - capacitors is strongly impacted by n , which decreas es as n s s increases .
Fig. 16 Super - capacitor maximum voltage Fig. 17 Super - capacitor energy capacity C. Integrated Model Test Based on the results presented to this point , a set of electrical system p arameter values were chosen for simulation with the integrated model . Those system parameters are listed in Table 4 . The amount of useful energy capacity was Table 4 . Integrated model parameters selected for testing Value Component Parameter Dual - Spool Dual - Spool HPS Config.
Config. v1 Config. v2 Super - Number of Super - Capacitors in Series 6 10 Capacitor Number of Super - Capacitor Strings in 12 15 Bank Parallel Useful Energy Capacity 0.8 kW - hr 2.5 kW - hr Bank Capacitance 100 F 75 F Minimum Bank Voltage 30 V 23 V Full Charge Bank Voltage 252 V 508 V Maximum Bank Voltage 303 V 564 V Total Bank Energy Capacity 1.27 kW - hr 3.31 kW - hr Super - Capacitor Capacitance 50 F 50 F Maximum Super - Capacitor V oltage 50 V 56 V Super - Capacitor Energy Capacity 0.018 kW - hr 0.022 kW - hr Maximum Super - Capacitor 174 A 177 A Continuous Current Charging Power 75 hp Super - Capacitor Peak Current 1900 A LPS EM Maximum Power 610 hp 410 hp N/A Maximum Torque 550 ft - lb 350 ft - lb N/A f f Maximum Torque Rate 150 ft - lb /s 90 ft - lb /s N/A f f HPS EM Maximum Power 750 hp Maximum Torque 200 ft - lb f Maximum Torque Rate 250 ft - lb /s f set to 0.8 kW - hr for the dual - spool configuration variants and 2.5 kW - hr for the HPS configuration. Note that the selected parameter values are for demonstration purposes and are not meant to represent a realistic physical system.
The test case w as chosen to be at a Mach number of 0 . 2 and an altitude of 5,000 ft . These conditions are consistent Fig. 18 HPC SM (a) and LPC SM (b) responses for a burst and cho p at a Mach number of 0.2 and an altitude of 5,000 ft Fig. 19 EM torque and power responses for a burst and chop at a Mach number of 0.2 and an altitude of 5,000 ft. HPS EM torque and power are plotted in (a) and (c), respectiv ely. LPS EM torque and power are plotted in (b) and (d), respectively.
- with an aborted landing scenario at a high altitude airport, which could be a potential worst case design scenario . The power level was increased from idle to ful l throttle over a duration of 1 s. After a brief time at the maximum power , the power was returned to idle over a period of 1 s. Figure 18 shows the stall margin variations compared to the baseline engine . Generally speaking, the minimum stall margins were increased . Note that the HPC S M is reduced during the charging period of the HPS configuration. Figure 19 show s the EM torque and power responses . The plots verify that the EM constraints were not exceeded. The power extraction observed after the dec eleration was for charging the super - capacitors. Finally Fig. 20 shows the super - capacitor bank voltag e and current to verify that the voltage and current limits were not exceeded. Note that the current plotted in Fig. 20 is the individual super - capacitor current, not the overall current going through the entire bank.
Fig. 20 Super - capacitor voltage (a) and current (b) responses for a burst and chop at a Mach number of 0.2 and an altitude of 5,000 ft V. Discussion In - direct and direct electrical system recommendations have been made for the implementation of the TEEM concept. Prima ry results include the suggestion for a 750 hp HPS EM with the ability to achieve maximum torque within 0.8 s. For dual - spool configuration v1, a 610 hp LPS EM is suggested with the ability to achieve the maximum torque within 3.7 s . The alternative varian t, dual - spool configuration v 2, allowed the LP S EM power to be r educed to 410 hp. Power need only be applied for the short duration of the transients, so it is possible that an undersized EM could be use d as long as the EM’s peak power capability is suffic ient. Furthermore, the analysis considered a fully deteriorated engine. Simulation data showed that , with the deterioration applied in the model, the EM requirements for a new engine were 30 - 40% lower than what is prescribed here. Based on this observation , trades could be made between component sizing and engine maintenance practices. Alternatively, it might be worth considering the use of different sized/powered EMs on the engines as it ages , or perhaps installing additional EMs on the spool(s) in order t o provide the additional required torque as the engine ages . It is also worthwhile to consider the variability of the system requirements in response to the deterioration model , and whether those variations are realistic .
While it is unclear how conservat ive the super - capacitor requirements will need to be, suggestions can be made based on the recommended power levels of the EMs and the energy required during a worst - case transient scenario.
Parametric analysis has provided data on the super - capacitor bank and individual cell or module parameters as well as the power extraction used for charging the super - capacitors. Conclusions drawn from the results are somewhat subjective and therefore specific suggestions are not provided. However, the results could be used to study trades in super - capacitor parameters that could drive development or selection in a given application. The analys is assumed that the individual super - capacitors were capable of handling a peak current of 1900 A based on the capabilities of pr esent day super - capacitors . If the peak current capability were more, its impact would generally be a reduction in the super - capacitor requirements. Because the ESD requirements are heavily driven by the EM requirement, the same arguments regarding deterio ration that applied to the EMs , also apply to the super - capacitors. In other words, significantly less energy storage would be required for a new engine than for an end - of - life engine, which implies potential trades between component sizing and maintenance practices. Once again, it is also worthwhile to consider if the deterioration model in this application is producing unrealistic variability in the engine performance that is producing overly conservative requirements.
There is an interesting trade off be tween the configurations. Dual - spool configuration v2 reduces the power requirement of the LPS EM by ~33% compared to dual - spool configuration v1. However, dual - spool configuration v1 allows for power extracted during transients to be used to charge the ES Ds, making more efficient use of the extracted power. While the HPS configuration eliminates the need for an LPS EM, it may require the super - capacitors to have significantly more energy storage because power is applied during both the acceleration and dec eleration transients. Thus there is a trade off between the HP S and dual - spool configurations. To help quantify th e trade off , 0.8 to 1.3 kW - hr of usable energy is suggested for the dual - spool configurations and 1.6 to 3 kW - hr of usable energy is sugge sted f or the HPS configuration, depending on how the worst - case transient scenario is defined . At minimum, the HPS configuration would trade a 410 or 610 hp EM for 0.8 to 1.7 kW - hr of super - capacitors. The energy storage would actually need to be larger than thi s considering the need to over - design for potential failure modes and to maintain a minimum bank voltage in order to prevent the current from exceeding its limits . Consider ing that the power density of EM s is expected to be about 9 kW/kg ( 5.5 hp/lb ) [ 18 ] and the energy density of super - capacitors is ~ 0.1 m kW - hr/kg ( 0.0045 kW - hr/lb ) and more typically ~0.05 kW - hr/kg (0.00225 kW - hr/lb ) , the trade off appears to be in m m favor of the dual - spool configuration. However, there is potential to drastically improve t he energy density of super - capacitors. Ref. [ 19 ] summarizes research that could potentially enable super - capacitors to achieve energy densities of up to 0.180 kW - hr/kg (0.0818 kW - hr/lb ) . Ref. [ 20 ] provides some data collected from other publications m on high performance conducting polymer - based super - capacitors that well surpass the capabilities of the current carbon - based super - capacitors. Being able to achieve these improvements in commercially available super - capacitors would substantially reduce system we ight and could very well shift the scales in favor of the HPS configuration .
It is very common for electrified aircraft concepts to include EMs and batteries , which has prompted a great deal of research in advancing the state - of - the - art in these types of devices. However, impulsive E SDs , such as super - capacitors , do not receive the same spotlight . Such devices could provide benefits and even play key roles in enabling electrified aircraft concepts. The ability to provide and absorb impulsive power could he lp to minimize the presence of batteries and reduce the overall system weight. It could help address the transition of aircraft modes of operation, particularly when considering vertical - lift vehicles . Of course, there is the obvious application to the TEE M concept as well. With potential uses in aircraft propulsion systems, a case can be made to advance the state - of - the - art in super - capacitors. Improving metrics such as energy density, energy capacity, peak current, and capacitance would promote the p racti cality of the TEEM concept and may go on to otherwise improve electrified aircraft concepts.
Application of the TEEM concept to a standalone gas turbine engine will add weight via EMs, ESDs, and other supporting equipment. Additional torque supplied by the EMs may requ ire the design of the shafts to be reconsidered, pot entially requiring a sturdier and heav ier design to withstand additional shear stresses . Furthermore, i mpacts on shaft inertia s could impact the electric system requirement s . While th ese are fair consideration s , the impact is expecte d to be small given that the maximum torque s supplied by the electric machines are less than ~3 % of the maximum torqu e applied to the turbines throug h the combustion of fuel . While TEEM will add weight to the propulsion system in some areas , it could reduce weight in o ther areas. For instance, if the HPS EM is capable of accelerating the HPS to a speed that is sufficient to draw in enough air into the engine to support combustion, then the HPS EM c ould double as a starter motor. This could allow for the elimination of t he APU , which adds weight on the order of 300 to 400 lb to the aircraft . Without an APU, there is no need for electric energy storage to support its starting. Thus those m energy storage devices could potentially be eliminated or used for implementing the T EEM concept . If the power of the EM is not sufficient for starting the engine by itself, it c ould be used to assist the engine start , thereby prompting a reduction in the size of the APU and its supporting ESDs . Alternatively , the APU could be replaced by an additional EM that can be used with the HPS EM to supply shaft power to enable engine starting . At the very least, the additional EM should be lighter than the APU it replaces . Even if an APU is present, the ESDs used to support it and other aircraft sy stems could serve a dual purpose . Those ESDs could be replaced or supplemented by the ESDs used in the application of TEEM. The operability benefits achieved by TEEM also create opportunities for design ing the engine in a way that improve s fuel efficienc y while also reducing the weight of the engine. Improvements in fuel efficiency could mean a reduction of onboard fuel storage which reduce s overall weight . The reduced weight of the propulsion system and fuel could enable an increase in the passenger/carg o capacity or an extension of an aircraft ’ s range . Further analysis is required to investigate the net benefit of implementing TEEM on the AGTF30 propulsion system, but the results to date seem to indicate that it has potential for enhancing the overall pe rformance of electrified gas turbine propulsions systems and the aircraft s they propel .
VI. Conclusion s An electrical system architecture was defined for evaluating the feasibility and practicality of the Turbine Electrified Energy Management (TEEM) concept a s implemented on a standalone turbofan. The electrical system has been modeled and integrated with an existing engine model that is representative of an advanced geared turbofan capable of producing 30,000lb of thrust at sea - level static conditions. The e ngine model incorporated health f parameters to simulate the impact of engine component degradation in order to assure the electrical system is capable of providing consistent and substantial operability benefits throughout the engine’s lifespan. Multiple co nfigurations for implementing TEEM were considered, including the use of electric machines (EMs) on both spools and use of a single EM on the high pressure spool (HPS). Regarding the use of EMs on both spools, two variants were considered that implemented different strategies for dealing with excess power during deceleration transients. One of those variants reduced the power requirement on the low pressure spool (LPS) EM by roughly 33%. While the configuration with a single EM on the HPS eliminates the LPS EM, it may also require more energy storage. The total amount of useful energy capacity recommended is 0.8 to 3 kW - hr , depending on the configuration and how the worst - case transient scenario is defined . The recommended peak power of the HPS EM was 750 hp . The recommend ed peak power of the LPS EM was 610 and 410 hp, depending on the variant of the dual - spool configuration. The EMs are required to achieve their maximum torque within 0.8 and 3.7 s for the HPS and LPS EM respectively . This is well within the capabilities of the current state - of - the - art. Elimination or dual - use of engine and aircraft equipment that is used to apply the TEEM concept will help to offset penalties for implementing TEEM. While further investigation is requir ed to validate the pract icality of implementing TEEM in this application, including the quantification of benefits, the r esults appear to be w ithi n the realm of possibility. It was conjectured that the net benefits of the TEEM concept could be further enhanced through advances in impulsive energy storage devices such as super - capacitors.
Acknowledgments The authors would like to acknowledge the Transformational Tools & Technologies (TTT) project under the A eronautics Research Mission Directorate (ARMD) that has supported this wor k. The authors would also like to acknowledge those within NASA who contributed to the creation of the AGTF30 model that was utilized in this study including Jeffryes Chapman and Scott Jones. Gratitude is also extended to Mark Bell of Vantage Partners LLC who provided consulting with regards to electrical system modeling .
References [ 1 ] Friedrich, C., and Robertson, P.A ., “Hybrid - Electric Propulsion for Aircraft,” Journal of Aircraft, Vol. 52, No. 1 (2015), pp. 176 - 189.
[ 2 ] Jansen, J.H., Bowman, C., Jankovsky, A., Dyson, R.., and Felder, J., “Overview of NASA Electrified Aircraft Propulsion Research for Large Subsonic Transports,” AIAA Propulsion & Energy Forum, Atlanta, GA. 2017.
[ 3 ] Hyun, D., Perry, A.T., and Ansell, P.J., “A Review of Distributed Electric Propulsion Concepts for Air Vehicle Technology,” AIAA Propulsion and Energy Forum, Cincinnati, OH. 2018.
[ 4 ] Johnson, W., Silva, C., and Solis, E., “Concept Vehicles for VTOL Air Taxi Operations,” Report Number ARC - E - DAA - TN50731. 2018.
[ 5 ] Antcliff, K.R., and Capristan, F.M., “Conceptual Design of the Parallel Electric - Gas Architecture with Synergistic Utilizat ion Scheme (PEGASUS) Concept,” 18th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, AIAA AVIATION Forum, Denver, CO. 2017.
[ 6 ] Culley, D., Kratz, J., and Thomas, G., “Turbine Electrified Energy Management (TEEM) For Enabling More Effic ient Engine Designs,” AIAA Propulsion & Energy Forum, Cincinnati, OH. 2018.
[ 7 ] Antcliff, K.R., Guynn, M.D., Marien, T.V., Wells, D.P.., Schneider, S.J., and Tong, M.T., “Mission Analysis and Aircraft Sizing of Hybrid - Electric Regional Aircraft,” AIAA 2016 - 1028, AIAA SciTech, 54th AIAA Aerospace Sciences Meeting, San Diego, CA, January 4 - 8, 2016.
[ 8 ] Thomas, G., Culley, D., Kratz, J., and Fisher, K., “Dynamic Analysis of the hFan, a Parallel Hybrid Electric Turbofan Engine,” AIAA Propulsion & Energy Forum, Cincinnati, OH, 2018.
[ 9 ] Kratz, J., Culley, D., Thomas, G., “A Control Strategy for Turbine Electrified Energy Management,” AIAA Propulsion & Energy Forum, Indianapolis, IN. 2019.
[ 10 ] Chapman, J., and Litt, J., “Control Design for and Advanced Geared Turb ofan Engine,” AIAA Propulsion & Energy Forum, Atlanta, GA, 2017.
[ 11 ] Chapman, J., Lavelle, T., May, R., Litt, J., and Guo, T. - H., “Toolbox for the Modeling and Analysis of Thermodynamic Systems (T - MATS) User’s Guide,” NASA TM - 2014 - 216638, 2014.
[ 12 ] Jone s, S., Haller, W., and Tong, M., “An N+3 Technology Level Reference Propulsion System,” NASA TM - 2017 - 219501, 2017.
[ 13 ] Claus, R.W., Evans, A.L., Lylte, J.K. and Nichols, L.D., “Numerical Propulsion System Simulation,” Computing Systems in Engineering, Vol 2, No. 4, pp 357 - 364, 1991.
[ 14 ] May, R.D., Csank, J., Lavelle, T.M., Litt, J. S., and Guo, T. - H., “A High - Fidelity Simulation of a Generic Commercial Aircraft Engine and Controller,” Proceedings of the 46th AIAA/ASME/SAE/ASEE Joint Propulsion Conference, A IAA - 2010 - 6630, Nashville, TN, July 2010.
[ 15 ] Chapman, J., and Litt, J., “An Approach for Utilizing Power Flow Modeling for Simulations of Hybrid Electric Propulsion Systems,” 2018 AIAA/IEEE Electric Aircraft Technologies Symposium, Cincinnati, OH, July 20 18.
[ 16 ] Bradley, M.K., and Droney, C.K., “Subsonic Ultra Gree n Aircraft Research: Phase II - Volume II – Hybrid Electric Design Exploration,” NASA/CR - 2015 - 218704/Volume II. April, 2015.
[ 17 ] “125V Heavy Transportation Module Datasheet”, Maxwell Technologies. URL: https://www.maxwell.com/images/documents/125V_Module_datasheet.pdf [retrieved 15 March 2019] [ 18 ] Sadey, D., Csank, J., Hanlon, P., and Jansen, R., “A Generalized Power System Architecture Sizing and Analysis Framework,” AIAA Propulsion & Energy Forum, Cincinnati, OH. 2018.
[ 19 ] “Alternative to traditional batteries moves a step closer to reality after exciting progress in super - capacitor technology” PHYS.ORG. February, 2 018.
URL: https://phys.org/news/2018 - 02 - alternative - traditional - batteries - closer - reality.html [retrieved 24 May 2019] [ 20 ] Shown, I., Ganguly, A., Chen, L.C., a nd Chen, K. - H., “Conducting polymer - based flexible supercapacitor,” Energy Science & Engineering , Society of Chemical Industry and John Wiley & Sons Ltd. 2014. URL: https://onlinelibrar y.wiley.com/doi/abs/10.1002/ese3.50 [retrieved 24 May 2019]