Integrated Design of a Variable Cycle Engine and Aircraft Thermal Management System Free

The continued modernization efforts of fifth-generation aircraft such as the F-35 Lighting II, in parallel with the development of sixth-generation aircraft concepts, has underscored the need for these aircraft to improve upon their maneuverability, range, stealth, and ability to handle high-powered electronics systems [1–4]. Advances in propulsion and thermal management system design are key to improving modern aircraft capability, according to Lockheed Martin [4]. Estimates for the required heat dissipation capability on these modern aircraft range from half a megawatt to over 15 MW [5,6], with waste heat primarily coming from electronics, weapons systems, and cockpit cooling.

Variable cycle engines, which enable improved thrust, fuel burn, and thermal management capability, are being developed as a response to the demands of advanced new aircraft [1,7,8]. Variable cycle engines are able to modulate the internal flowpaths inside the engine in order to operate in high-thrust or fuel efficient modes, depending on the needs of the aircraft at a given point in the mission.

It has been demonstrated that there is benefit to integrating the design of the propulsion system with the aircraft thermal management system, with Maser developing an analysis method based on the minimization of exergy destruction [9]. More recently, Jasa examined coupled design of a fuel-cooled TMS and a simple closed-loop air cycle machine (ACM) in the context of trajectory optimization for a supersonic aircraft utilizing a simplified three stream engine with limited variable geometry features [10].

Allison et al. [11] used a three-stream VCE developed by Simmons [12] to examine freestream Mach-dependent performance at a single altitude for an open-loop ACM that was sized at a single operating condition.

Existing literature has highlighted the need to look at overall mission performance [13], instead of looking at the propulsion system and thermal management system in isolation or examining only a small subset of operating points in the mission. As such, the primary objective of this investigation is to develop an integrated model of a variable cycle engine and notional thermal management system architecture that can be used to examine coupled performance of the propulsion system and TMS across a complete flight envelope. The VCE model is intended to provide performance similar to an F-35 class powerplant, but the goal of this work is not to attempt to optimize performance of the propulsion system. Rather, the goal is to use the coupled VCE-TMS model to elucidate the importance of the integrated design of these two critical aircraft subsystems. Furthermore, key aircraft-level design trades associated with the selection of the thermodynamic design and sizing point of the thermal management system will be explored, as the selection of the design point can lead to significant differences in off-design behavior.

Figure 1 shows a schematic of the variable cycle engine under investigation in this study. The VCE is an augmented double-bypass, three-stream engine, which can be thought of as a separate-flow turbofan (SFTF) wrapped around a mixed-flow turbofan (MFTF). The variable geometry components in the engine include variable inlet guide vanes (IGVs) on each compressor, a variable area bypass injector (VABI) leading into the inner bypass/core mixer, and two variable throat nozzles. Modulation of these variable geometry features allows for flow to be shifted away from the core stream and into the inner bypass and third stream. As Simmons [12] notes, moving more flow into the bypass streams as the engine is throttled down allows for a reduction in inlet spillage drag, which can lead to a significant improvement in specific fuel consumption (SFC). However, Simmons and Clark et al. [12,14] both showed that excessive variable geometry modulation can lead to extremely high Mach numbers in the two bypass ducts. At a certain point, the pressure losses associated with these high Mach numbers overcome the benefit associated with the reduction in spillage drag.

Clark et al. [14] provide an in-depth description of the modeling approach used for this architecture, so only a brief overview will be given here. A zero-dimensional thermodynamic model of the VCE is built in the numerical propulsion system simulation (NPSS) tool, the industry standard for modeling air-breathing propulsion systems. NPSS solves a set of equations using a Broyden-modified Newton–Raphson solver to ensure that energy, mass, and momentum are conserved in the propulsion system model [15,16].

The multiple design point (MDP) method proposed by Schutte [17–19] is used to size the engine to achieve design targets at four different points, including an aerodynamic design point (ADP), Top of Climb (TOC), Takeoff (TKO), and Sea Level Static (SLS), summarized in Table 1. At the ADP, the engine is sized for a certain amount of corrected flow passing through each component and physical flow areas are defined. In addition, turbomachinery map scalars are set, which then define how the turbomachinery components perform at all other off-design operating conditions [16]. The TOC point, which occurs at the same flight conditions as ADP, is typically used to specify a thrust target so that a required rate of climb can be achieved at the top of climb. The difference between ADP and TOC is that schedules for the variable geometry features are enabled at TOC, while they are not enabled at ADP. For this study, the engine is set to run to a temperature target rather than specifying a thrust target at TOC. The TKO point is where the turbine inlet temperature (T₄₁) is the hottest, so cooling flow fractions must be set at the ADP in order to achieve the required turbine cooling flow at TKO. The SLS point has a rated thrust target that must be met for the aircraft to accelerate down the runway, so the engine airflow at the ADP must be sufficient for the SLS point to achieve the required thrust. Both the TKO and SLS points were evaluated at elevated temperatures (+27 °F) relative to the International Standard Atmosphere (ISA) to simulate the need for performance on hot days. Furthermore, although the VCE architecture has an afterburner present, none of the MDP points utilize augmented thrust. The afterburner is only enabled during off-design evaluation of the complete flight envelope, when maximum wet thrust is set by limiting the fuel to air (FAR) ratio entering the afterburner to 0.045.

The MDP method uses two solvers within the NPSS model. An outer MDP solver varies the design parameters for the engine at the ADP, and checks off-design performance parameters at the other design points. An inner cycle (CY) solver is used at each design point to ensure the conservation equations are satisfied. The MDP solver iterates the CY solver through each design point to ensure that the design requirements are all met simultaneously.

The thrust target used by Clark et al. [14] is maintained in this study. The VCE is sized to achieve SLS dry thrust performance similar to that of the Pratt and Whitney F-135 engine used in the Lockheed Martin F-35 Lightning II [20]. For this engine modeling effort, a maximum T₄₁ value of 3350 R (1861 K) is selected. Public domain information indicates that the F-135 can achieve turbine inlet temperatures of 3600 °F (4060 R, 2255 K) [21], so a maximum T₄₁ of 3350 R is well within demonstrated capabilities of modern military aircraft without requiring state of the art turbine materials and cooling technology. The goal of this engine model was to achieve similar thrust performance, not to match the exact cycle parameters of the F-135. For ambient conditions above Mach 0.4 (i.e., beyond takeoff and initial climb), the maximum T₄₁ was derated to 3200 R (1778 K) to simulate operational constraints intended to improve component durability during cruise. For the selected SLS operating condition (maximum T₄₁), the engine operates to the right of the theta break, so it is not operating simultaneously at the maximum OPR and T₄₁ at SLS. The key design parameters for the engine are shown in Table 2.

This VCE model contains three key types of variable geometry: variable IGVs, a rear VABI at the inner bypass/core mixing plane, and variable throat nozzles. The effect of variable IGVs is handled in the performance maps used to define compressor performance. CMPGEN [22], which is a joint industry/NASA tool developed to estimate compressor performance, is used to produce compressor maps that contain no variable IGV effects. The base maps produced by CMPGEN are then modified to account for IGV position based on a reference variable geometry compressor map. The reference map is used to create scale factors based on compressor corrected speed and IGV angle that adjust the compressor corrected flow, pressure ratio, and efficiency as the inlet guide vane is closed down. The process is explained in more detail by Clark et al. [14].

At the ADP, the design nozzle throat area is calculated based on the pressure ratio across the nozzle. In off-design, the nozzle throat area can be changed as an independent input to the model. The ratio between the off-design throat area and the design throat area is tracked and limited to prevent excessive throat variation, as suggested by Simmons [12].

The TMS used in this study is nearly identical to the environmental control system (ECS) architecture initially proposed by Shi et al. [24] and modified by Clark et al. [13] for use as an aircraft thermal management system. The key difference here is that the primary and secondary heat exchangers are placed in the VCE third stream, as shown in Fig. 1, rather than in a separate ram air stream. A brief description of the architecture is provided here, with a more complete description available from Clark et al. [13].

Figure 2 shows a simplified schematic of the open loop air cycle system (ACS) linked to the variable cycle engine, where the arrows indicate the direction that air flows through the system. This ACS architecture is similar to a standard ECS architecture, with the primary difference being that the goal of the ACS is to transfer heat away from the aircraft, while an ECS is intended to provide cool air to the aircraft cabin. Pressurized working air is bled from the VCE high-pressure compressor (HPC), and is then cooled in a primary heat exchanger placed in the third stream. The cooled air is then pressurized further in an ACS compressor (ACSCmp1), before passing through a secondary heat exchanger also located in the engine third stream. The working air then passes through a water extraction loop, intended to cool the air down sufficiently for water vapor to condense and be separated from the flow. After the water extraction loop, the air is expanded to a low temperature in two cooling turbines (ACSTrb1 and 2). Removal of water vapor is important given that the two cooling turbines reach subfreezing temperatures, and formation of ice inside the turbines could damage them.

After exiting the cooling turbines, the cold ACS working air enters a heat exchanger connected to a thermal transport bus (TTB). The TTB is a small loop that has a heat input element to represent the heat load rejected from the aircraft internal cooling system. Depending on the fluid used in the TTB, either a pump or small compressor would be required to circulate the fluid through the loop, which is not depicted in Fig. 2. The ACS working air exits the TTB heat exchanger and is simply dumped overboard. To maximize heat transfer capability, the air is set to expand out to nearly atmospheric static pressure, so it has negligible thrust potential when it exits the TTB heat exchanger. The TTB heat exchanger bypass valve (HXByp) was shown to have some potential use by over sizing the TMS in order to improve off-design performance in some situations [13], but it is not used in this study.

The turbine bypass valves depicted in Fig. 2 are used to control temperatures within the ACS. The first turbine bypass valve (Trb1Byp) can be used to reduce the overall work available to the ACS compressor, which may be necessary if the temperature at the exit of ACSCmp1 reaches material limits. The second turbine bypass valve (Trb2Byp) can be used to control the temperature of the ACS working air that enters the TTB heat exchanger. Some potential heat transfer fluids, such as Dow SYLTHERM 800 [25] have a minimum operating temperature (−40 °F/C in the case of SYLTHERM 800), so the temperature of the working air that cools the TTB fluid in the TTB HX must be regulated. If air is used in the TTB loop, no such temperature restriction is necessary. In their prior study of this TMS architecture, Clark et al. [13] showed that the choice between air or a heat transfer fluid such as Dow SYLTHERM 800 had no impact on the thermodynamic cycle of the ACS, but could potentially impact the design of the TTB heat exchanger. In that study, the minimum operating temperature of SYLTHERM (−40 °F/C) was applied as a limit in the TTB loop. In this study, it is assumed that air is used in the TTB loop, so the minimum TTB temperature restriction has been lifted. The impact of lifting this temperature constraint is significant and will be discussed further.

The most significant difference between this investigation and the prior work conducted by Clark et al. [13] is the integration of the thermal management system into the engine MDP phase. In the prior study, the low bypass ratio mixed-flow turbofan was designed at a single point, SLS, and then the thermal management system was subsequently designed while the engine was operating in off-design mode. It is important to re-emphasize the meaning of the word design. The design of the engine refers to the physical sizing of flow areas and rates at a given flight condition, as well as the calculation of turbomachinery map scalars. Off-design operation evaluates the performance of the sized engine at any ambient flight condition. Similarly, when the design of the TMS is discussed, it refers to the sizing of the components in the TMS.

Herein lies the advantage of the MDP method: by incorporating the TMS into the MDP phase, the MDP solver in NPSS will ensure that the reduction in thrust associated with operating the TMS is offset by sizing the engine with enough total airflow to meet the required thrust target(s). This does not eliminate any performance impact to the engine however, as the turbines must still provide work to compress the air that is eventually bled off into the TMS. To limit the negative performance impact to the engine, the maximum allowable bleed flow from the VCE to power the TMS is set to 10% of the high-pressure compressor (HPC) inlet mass flow, referred to as W₂₅.

A significant drawback noted by Clark et al. [14] was the difficulty in incorporating a gradient-based optimizer into the constrained NPSS cycle solver that is used to calculate off-design performance. As such, a pattern search method, adapted from that described by Nocedal and Wright [27], is developed for this study to perform optimization on the variable geometry features. The pattern search method works by starting at a given position for a set of k design variables $X$⁠. Each design variable is independently stepped by some amount $±ΔXk$⁠, and the value of the objective function f is noted. The point $X±ΔXk$ associated with the largest improvement in the objective function f is then used as the next starting point, and the process is repeated until no improvement in the objective function is found. If no improvement in the objective function is found for a given iteration, the step size $±ΔXk$ is first reduced by 25%, and the objective function is reevaluated for each $X±ΔXk$⁠. The step size is allowed to reduce five times before the objective function is deemed to be sufficiently optimized due to a lack of improvement along the direction of any of the design variables. A fractional tolerance is also applied, and if the objective function improves by an amount less than the tolerance, the pattern search method is terminated.

The variable geometry component positions serve as the independent design variables $X$ for this optimization method, while engine specific fuel consumption (SFC), maximum heat dissipation capability Q, and net thrust F_n serve as the potential objective functions. The pattern search optimizer, while effective at finding optimal variable geometry settings, is not particularly efficient from a computational standpoint. Therefore, the optimizer is used to build tabulated schedules of variable geometry positions versus engine pressure ratio. NPSS is extremely fast at interpolating one-dimensional tables, so for efficient off-design evaluation, the method of tabulated schedules works quite well.

Unlike the version of the VCE used by Clark et al. [14], the engine used in this study has the afterburner enabled for augmented military thrust operation. When the afterburner is on, the core nozzle area must naturally increase to account for the lower density of the high-temperature exhaust flow. As such, directly scheduling nozzle throat area would require separate schedules for augmented versus nonaugmented operation. Instead, the fan and LPC operating lines are the target parameters to be scheduled, with the nozzle throat areas being varied to achieve the target operating line.

During augmented operation, the engine core will always be operating at its maximum power setting. Since the effect of adding fuel to the exhaust takes place downstream of all turbomachinery components, the core engine will not notice any appreciable difference between augmented and maximum power nonaugmented operation. This is because the engine pressure ratio is high enough to keep the nozzle choked at all times during maximum power operation, even with any pressure loss associated with the afterburner. Therefore, when operating with the afterburner on, the variable geometry schedules are maintained at their maximum power settings.

To run the engine through a power hook (also known as a throttle sweep) at any given ambient flight condition, the optimizer is first used to set the variable geometry to produce the maximum possible dry (nonaugmented) thrust. This thrust level is often referred to as intermediate rated power (IRP). The IRP variable geometry settings are then used throughout the wet (augmented) thrust regime. The separately optimized and tabulated variable geometry settings are used as the engine throttle is reduced from maximum dry thrust down to minimum idle thrust.

As stated previously, the purpose of this investigation is to extend prior work [14] and develop a better understanding of how the proposed TMS architecture interacts with a variable cycle engine. Furthermore, it is desired to determine how the selection of the TMS design point and design heat load impacts overall TMS heat dissipation capability. The benefit of integrating the TMS design into the engine design process will also be discussed. Results will be presented both in the context of TMS performance at select operating conditions, as well as overall aircraft mission-level performance.

One key metric of interest is the maximum waste aircraft heat that the proposed ACS-style TMS is capable of dissipating. As alluded to previously, there is the potential for a minimum temperature constraint on the fluid in the thermal transport bus, depending on the selection of that fluid. The presence of this temperature constraint turns out to have a significant impact on the maximum heat dissipation capability of the system, as well as the optimum design point for the TMS.

Two different design points are investigated for the TMS: a sea level static (SLS) hot day point, and the engine aerodynamic design point (ADP). These points are chosen because they represent roughly opposite corners of a typical subsonic flight envelope. For both the ADP and SLS design point, the integrated VCE/TMS model is exercised to determine the maximum heat dissipation capability of the TMS. This is done with the TTB minimum temperature constraint enabled and disabled.

It should be noted that at the ADP, all variable geometry features are by definition at their design settings, so the engine cannot attempt to modulate the geometry to handle a higher heat load. However, the nature of the variable geometry modulation is such that it can be varied to allow the TMS to handle higher heat loads. Therefore, TOC maximum heat dissipation capability is also reported, since the variable geometry schedules are enabled at the TOC design point within the MDP setup. Table 3 shows a summary of the maximum heat dissipation capability of the TMS when the TTB loop has no minimum operating temperature constraint. The maximum heat dissipation capability is set by restricting HPC bleed flow going to the TMS to be 10% of HPC inlet flow W₂₅. For reference, total turbine cooling flow (split between the high-pressure and low-pressure turbines) is approximately 17% of W₂₅.

The effect of the variable geometry is clear from Table 3, which shows that the maximum heat dissipation at TOC is significantly higher than the maximum capability at ADP, where the variable geometry schedules are not enabled. In addition, it would appear that designing the TMS at ADP conditions appears to lead to higher heat dissipation capability across the board. However, simply looking at the maximum heat dissipation capability does not tell the entire story. Figure 3 contains a plot of maximum heat dissipation capability versus thrust for each TMS design point at top of climb conditions. Figure 3 shows that while designing the TMS at ADP leads to a higher maximum heat dissipation capability, the SLS design point has higher heat dissipation capability at lower thrust settings.

Table 4 shows a summary of the maximum heat dissipation capability when the thermal transport bus minimum temperature constraint is activated. Comparison of Table 4 and Table 3 shows a significant reduction in heat dissipation capability when the minimum temperature of the TTB fluid is constrained. This demonstrates that using a heat transfer fluid that cannot reach extremely cold temperatures is quite detrimental to TMS performance.

To demonstrate the benefit of integrating the TMS into the engine design process, Table 5 shows the maximum heat dissipation capability for the TMS architecture when it is designed with an unconstrained TTB temperature after the engine design has been completed, the same approach taken by Clark et al. [13]. Comparing the maximum capability for this separate design process to the integrated process results in Table 3 shows a notable decrease in maximum heat dissipation capability. This reduction in capability is primarily because the HPC is not designed for the level of bleed flow required to power the TMS. However, a reduction in TMS performance is not the only downside to the separate design process.

Tables 6 and 7 show the maximum IRP thrust (maximum nonaugmented thrust) that the engine can achieve while attempting to maintain as much heat dissipation capability as possible. There is a significant decrease in maximum available thrust that the engine can produce when the TMS is designed separately from the engine. This is due the separate design process producing an undersized engine core, since the design point for the core does not account for the TMS bleed flow.

The prior work conducted by Clark et al. [13] used a MFTF engine as the base engine. In one model, the primary and secondary TMS heat exchangers were placed in a separate ram air stream, while a second model placed the heat exchangers in the MFTF bypass stream. It is worth providing a direct comparison of the heat dissipation capability of the prior MFTF models versus the integrated VCE and TMS model. Figure 4 shows the results of a sea level static power hook when the TMS is designed with a 50 Btu/sec (52.8 kW) heat load. The x-axis is scaled by the maximum net thrust to account for the differences in thrust produced by the two different engine designs. The y-axis shows the percentage of the design heat load that can be maintained as the engine is throttled down. The benefit of the variable cycle engine is clear from Fig. 4. The TMS, when integrated with a VCE, is able to dissipate a higher percentage of the design heat load across the entire power hook.

Another key advantage of the variable cycle engine is the ability to optimize the geometry for different aircraft mission figures of merit. The pattern search optimizer is used to generate variable geometry schedules that would maximize thrust, maximize heat dissipation capability, or minimize engine specific fuel consumption. The prior results all used the variable geometry schedules intended to maximize heat dissipation. To demonstrate the difference, Fig. 5 shows the HPC inlet guide vane schedules developed with the optimizer for the three different objective functions.

Figure 6 shows the difference in heat dissipation capability for a power hook at top of climb conditions for the two variable geometry schedules, while Fig. 7 shows the difference in specific fuel consumption at the same conditions. Figures 6 and 7 clearly demonstrate the utility of the variable cycle engine: if the aircraft is operating at a condition where it needs higher heat dissipation capability from the thermal management system, the variable geometry can be modulated to provide that extra capability. If, on the other hand, the aircraft is cruising and would benefit from lower fuel consumption, the variable geometry schedule can be swapped to reduce fuel burn.

Table 8 shows a comparison between pertinent cycle parameters and variable geometry settings at top of climb conditions (35,000 ft, Mach 0.8) for the max Q versus min SFC schedules. The value of each parameter at the cycle design condition ADP (also 35,000 ft and Mach 0.8) is also shown for reference. Recall that at ADP, the mixer and nozzle throat areas are being sized and the IGV angles are set to 0 deg. Examination of Table 8 shows that to maximize heat dissipation capability, flow to the inner bypass stream is reduced, which sends more flow to the core. Flow to the third stream is increased, since the TMS heat exchangers are located in the third stream.

A comparison of the SFC values in Table 8 between ADP and the minimum SFC schedule at TOC shows that ADP actually achieves lower SFC than when the variable geometry is modulated to reduce SFC. While this result may appear counterintuitive, since the premise of modulating the geometry is that it can improve SFC, note that the TOC setting achieves higher heat dissipation and thrust than at ADP. Lowering the demanded heat load and thrust target at TOC to match the ADP values while using the minimum SFC schedule resulted in slightly improved SFC (0.920 lbm/hr/lbf) relative to ADP. Running the pattern search optimizer at this single point further reduced SFC down to 0.919 lbm/hr/lbf.

A re-inspection of Table 6 shows that the maximum available SLS thrust is greater than the 28,000 lbf thrust target used in the MDP design phase, from Table 2. It is the difference in engine performance driven by the different variable geometry schedules that creates this difference in design thrust versus realized thrust. During the MDP phase, the engine was designed with the variable geometry schedules set to maximize heat dissipation capability. When determining the maximum thrust values in Table 6, the variable geometry schedules are switched to the schedules developed by the optimizer to maximize thrust.

The prior results are useful to demonstrate the capability of the thermal management system at selected points in the aircraft operating envelope. However, as Clark et al. [13] concluded, an examination of aircraft mission-level results is needed to provide more insight into the optimal design point for the TMS.

NASA's Flight Optimization System (FLOPS) [28] is a tool that can be used to simulate aircraft performance across an entire mission. FLOPS was used to simulate a fixed-fuel F-35 flying a simple climb-cruise-descent mission, where FLOPS selected the best cruising altitude for maximum range. Engine performance decks are generated using the integrated VCE/TMS model and fed into FLOPS to determine the differences in range that would arise from the different variable geometry schedules, TMS design points, and design heat loads. It should be noted that the only differences in FLOPS inputs are the engine decks and the maximum sea level static thrust. Although marginal differences in engine diameter and weight would arise from the slightly different design airflow values of the various engines, calculation of these differences was not included in this study. A more detailed aircraft-level analysis would account for these differences, but for first-order analysis, maintaining a constant engine weight and diameter will provide useful notional trends.

Table 9 shows the FLOPS-calculated design range for both TMS design points (ADP and SLS), as well as for two different variable geometry schedules (maximizing heat dissipation Q or minimizing engine SFC). Also shown is the range for a reference VCE using the SFC minimizing variable geometry schedule that was designed to the same performance parameters from Table 2, but has no thermal management system attached. For the other engines in Table 9, the TMS was designed for a 100 Btu/sec (52.8 kW) heat load. The effect of the variable geometry schedules is quite clear, with the SFC-minimizing schedules showing an increase of over 300 nautical miles in range versus the Q-maximizing schedules. The range values are quite similar regardless of the TMS design point. Furthermore, the impact of incorporating the TMS on aircraft range is evident when comparing range values to the reference VCE.

Table 10 shows range values for when the TMS is designed to the maximum heat dissipation capability. Once again, there is a notable improvement in range when the SFC-minimizing variable geometry schedule is used. Unlike the 100 Btu/sec design load, however, when designed to the maximum heat dissipation capability, range is notably improved when the TMS is designed at the ADP. In addition, the higher design heat load of the TMS reduces the maximum range significantly, compared to the reference VCE with no TMS attached.

However, recall from Fig. 3 that the SLS design point has higher heat dissipation capability at all but the highest thrust settings. While the ADP design point may provide better range, the SLS design point provides better heat dissipation capability at most off-design operating conditions. Examining the heat dissipation capability at each point in the mission profile from the FLOPS output shows that the SLS design point provides better heat dissipation capability throughout the mission. Figure 8 shows the heat dissipation capability of the TMS for the two design points in the cruise portion of the notional mission plotted versus distance flown, for the case when the TMS was designed for maximum heat dissipation capability. Note that heat dissipation capability slowly decreases as fuel is burned off and the engine is slowly throttled back throughout the cruise.

A notional supersonic combat mission was developed to simulate a more realistic mission with varying operating conditions. In the combat mission outbound leg, the aircraft climbs to 30,000 ft into an ingress cruise at Mach 0.8 and then loiters for 60 min. For the inbound leg, the aircraft accelerates from Mach 0.8 to Mach 1.6 and performs a 360 deg turn to simulate combat, followed by a 50 nm dash at Mach 1.6. The aircraft then decelerates to Mach 0.85 for the egress cruise home. FLOPS was set to adjust the ingress and egress cruise distances to match distance for the outbound and inbound legs of the mission.

Table 11 shows the total range (twice the mission radius), along with the average heat dissipation $Q¯$ capability of the TMS during each mission segment. Average heat dissipation is reported since the engine power is slowly adjusted by FLOPS during each segment as aircraft weight changes, leading to a change in heat dissipation capability, as seen in Fig. 8. During high-power operation, such as the Mach 1.6 turn and dash, the TMS designed at ADP conditions is able to achieve higher heat dissipation capability. Once again, at the lower power cruise conditions, the TMS designed at SLS conditions provides better heat dissipation capability. Although the ADP-designed TMS has a slightly higher range, the difference in range is small primarily because a significant portion of the mission fuel is spent loitering, for which no range credit is given.

It should be noted that the engine was allowed to operate up to the maximum T₄₁ of 3350 R at all freestream Mach numbers when building the engine performance deck used in FLOPS for the combat mission, unlike the climb-cruise-descent mission where up and away performance was limited to 3200 R.

These results demonstrate that the selection of the TMS design point greatly depends on the aircraft mission parameters. There is a clear tradeoff between range, heat dissipation capability, and engine power setting. For a simple climb-cruise-descent mission where the engine is primarily operating at a part-power cruise condition, the higher heat dissipation capability provided by the SLS design point may be preferred if heat dissipation capability is the primary concern. However, a mission that involves high power combat maneuvers may prefer a TMS designed at the ADP, given the higher heat dissipation capability of an ADP-designed TMS at high power settings.

An integrated variable cycle engine and thermal management system model was developed that is capable of providing a propulsion system designer with insight into how various design choices can impact overall propulsion and aircraft system capability. The impact of the TMS design point on overall heat dissipation capability was demonstrated, and the relationship between heat dissipation capability and engine power reveals the tradeoffs that arise from the selection of the TMS design point. Systems designed at ADP conditions tend to perform better at high power settings with extended cruise range, while systems designed at SLS conditions tend to have higher heat dissipation capability when the engine is operating at part power conditions. This demonstrates the importance of the aircraft mission when designing the propulsion and thermal management systems.

The VCE/TMS model was also used to highlight the need for integrated design of the thermal management system within the engine design process. It was shown that if the two systems are not designed together, the resultant propulsion system will be significantly undersized, producing lower thrust and capable of dissipating less heat from the aircraft. Furthermore, when compared to prior studies that examined the same TMS connected to a fixed-cycle MFTF, the VCE was shown to have improved heat dissipation capability.

The impact of the variable geometry schedules was also demonstrated. The variable geometry features in a VCE provide greater flexibility in the design and operation of an engine compared to a fixed-cycle engine, allowing the propulsion system designer to target different aircraft mission figures of merit. It was shown that different schedules can be developed to improve thrust, heat dissipation capability, or fuel burn.

Future studies would benefit from integration of propulsion system weight estimation tools. In addition, the current modeling efforts required manual execution of FLOPS, which makes it difficult to perform system-level optimization studies. Future work would stand to benefit from coupling the propulsion system, weight, and aircraft analysis into an overall system design loop, as demonstrated by Kirby and Mavris [29] and Clark et al. [30]. Lastly, this study did not explore optimization of the variable cycle engine itself. There is undoubtedly room for optimization of the design parameters for the engine (pressure ratios, temperatures, bypass ratios, etc.) when modeled with the coupled TMS.

Robert Clark was supported in this research by the Department of Defense (DoD) through the National Defense Science and Engineering Graduate (NDSEG) Fellowship Program.

• Office of Naval Research (Award No. NDSEG Fellowship; Funder ID: 10.13039/100000006).

The datasets generated and supporting the findings of this article are obtainable from the corresponding author upon reasonable request.

ACS =

air cycle system

ACM =

air cycle machine

ADP =

aerodynamic design point

ECS =

environmental control system

EPR =

engine pressure ratio

FAR =

fuel to air ratio

HPC =

high-pressure compressor

HX =

heat exchanger

IGV =

inlet guide vane

IRP =

intermediate rated power

ISA =

international standard atmosphere

LPC =

low-pressure compressor

LPT =

low-pressure turbine

MDP =

multiple design point

MFTF =

mixed-flow turbofan

NcPct =

corrected speed as a percentage

Noz070 =

core nozzle

Noz270 =

third stream nozzle

NPSS =

numerical propulsion system simulation

OPR =

overall pressure ratio

SFC =

specific fuel consumption

SFTF =

separate-flow turbofan

SLS =

sea level static

TKO =

takeoff

TMS =

thermal management system

TOC =

top of climb

TTB =

thermal transport bus

VABI =

variable area bypass injector

VCE =

variable cycle engine

AqA =

mixer core area/bypass area

A_th =

throat area

F_g =

gross thrust

F_n =

net thrust

T₄₁ =

turbine inlet temperature

W =

mass flow

W₂₅ =

HPC inlet flow