Article

Design and Evaluation of a Hypersonic Waverider Vehicle

Using DSMC
Angelos Klothakis and Ioannis K. Nikolos *

Turbomachines & Fluid Dynamics Laboratory, School of Production Engineering and Management,
Technical University of Crete, 73100 Chania, Greece; aklothakis@gmail.com
* Correspondence: inikolos@tuc.gr

Abstract: This work investigates the aerodynamic performance of a hypersonic waverider designed
to operate at Mach 7, focusing on optimizing its design through advanced computational methods.
Utilizing the Direct Simulation Monte Carlo (DSMC) method, the three-dimensional flow field
around the specifically designed waverider was simulated to understand the shock wave
interactions and thermal dynamics at an altitude of 90 km. The computational approach included
detailed meshing around the vehicle’s critical leading edges and the use of three-dimensional iso-
surfaces of the Q-criterion to map out the shock and vortex structures accurately. Additional
simulation results demonstrate that the waverider achieved a lift–drag ratio of 2.18, confirming
efficient aerodynamic performance at a zero-degree angle of attack. The study’s findings contribute
to the broader understanding of hypersonic flight dynamics, highlighting the importance of precise
computational modeling in developing vehicles capable of operating effectively in near-space
environments.

Keywords: hypersonic vehicles; direct simulation Monte Carlo; aerodynamic performance;

computational fluid dynamics; hypersonic waverider

1. Introduction
The waverider is a distinct configuration of a hypersonic vehicle engineered to
maintain the shock wave generated by the vehicle attached to its leading edge, offering
Citation: Klothakis, A.; Nikolos, I.K.
numerous advantages [1]: (a) this shock wave effectively separates the flow fields on the
Design and Evaluation of a Hypersonic
upper and lower surfaces of the vehicle, thus preventing high-pressure flow from spilling
Waverider Vehicle Using DSMC.
from the lower to the upper surface at the leading edge; (b) it ensures a uniform flow on
Computation 2024, 12, 140. https://
the lower surface, which is optimal for scramjet engine intake; (c) the design of the
doi.org/10.3390/computation12070140
waverider is derived through reverse engineering from a known flow field around a
Academic Editor: Jeffrey S. Marshall conical shape. This design approach will be elaborated on in later sections of this work.
Received: 17 May 2024 The concept of the waverider was first put forward by Nonweiler in 1963 [2]. At that
Revised: 24 June 2024 time, the available technology was not advanced enough to produce a hypersonic air-
Accepted: 3 July 2024 breathing vehicle. Sandlin and Pessin offered an in-depth explanation of the design
Published: 9 July 2024 methodology for waveriders [1]. A broad analysis of the significant design challenges
associated with hypersonic flight vehicles is discussed in the work of Sziroczak and Smith
[3], where two types of vehicles, hypersonic transports and space launchers, are
Copyright: © 2024 by the authors. examined. Although these are distinct from waveriders, they share many design and
Licensee MDPI, Basel, Switzerland. construction challenges. Zhang et al. employed the three-dimensional Class–Shape
This article is an open access article Function Transformation (CST) method to determine the geometric shapes of Hypersonic
distributed under the terms and Gliding Vehicles (HGVs) [4]. Inviscid modeling methods, relying on modified Newton
conditions of the Creative Commons theory and the Prandtl–Meyer equation, were used to assess the aerodynamics of these
Attribution (CC BY) license vehicles. The use of the NSGA-II multi-objective optimization algorithm helped achieve
(https://creativecommons.org/license designs with high lift–drag ratios and substantial volumetric efficiency. In his Ph.D. thesis,
s/by/4.0/). K. Kontogiannis considered the parameterization and manipulation of waverider

Computation 2024, 12, 140. https://doi.org/10.3390/computation12070140 www.mdpi.com/journal/computation

Computation 2024, 12, 140 2 of 24

forebody geometries, proposing a new three-dimensional leading-edge waverider design

method, shifting from inverse to direct design methodologies [5]. Additionally, Son et al.
described an optimization framework for the inverse design of hypersonic waveriders [6].
Since its introduction, the waverider has been a focal point of research. After
conceptualizing the design, the next phase involves its analytical assessment. Various
methods such as engineering correlations, flight tests, wind tunnel tests, and numerical
simulations are employed not only for analysis but also for validating concepts and
accumulating knowledge. While effective in the subsonic domain, these methods have
shown limitations in reliability and implementation at hypersonic speeds. Ongoing
research aims to refine these methods to master the hypersonic domain.
During the early quest for hypersonic flight, NASA designed, constructed, and tested
a hypersonic vehicle [7]. The vehicle demonstrated that it was capable of achieving speeds
up to Mach 8. However, a closer look at wind tunnel data revealed an interesting
phenomenon. It was observed that the vehicle’s aerodynamic performance parameter lift–
drag ratio (L/D) decreased as Mach number increased. Kuchemann documented a
decreasing trend in maximum L/D with increasing Mach numbers through extensive
empirical research [8], highlighting a significant challenge in traditional aerodynamic
design and methodology in the hypersonic regime. Moreover, Bowcutt introduced a
multidisciplinary approach to optimize waverider configurations, achieving relatively
higher L/D ratios at higher Mach numbers [9], a finding later corroborated by other
researchers [10,11]. Nonweiler originally developed the waverider design technique in the
1950s to create a hypersonic wing for re-entry purposes [12]. Despite early challenges due
to the inherent sharpness of the waverider’s leading edge, which led to significant
aerothermodynamic loads, and the limited volumetric efficiencies of initial thin
configurations [7] that did not account for viscous effects, advancements in numerical
methods and computational resources in the 1980s enabled Bowcutt to optimize
waveriders by predicting the viscous effects on their performance [9]. This has spurred
continued interest and research into waverider designs, although practical application in
producing operational hypersonic vehicles remains a challenge.
Since Nonweiler [2] introduced the waverider design concept, leveraging wedge flow
dynamics to create the first waverider in aerospace science, numerous design studies have
aimed to refine and optimize his initial concept. The majority of these studies model the
flow as two-dimensional and inviscid, either planar or axisymmetric [13–22]. A smaller
subset has considered three-dimensional inviscid flow, both without and with chemical
reactions, as in references [23–28] and [29,30], respectively. Only a limited number of
investigations have addressed three-dimensional viscous flow, incorporating turbulence
[31]. This study focuses on analyzing the full three-dimensional viscous flow field around
the designed waverider.

2. Waverider Design Methodology

Waveriders are crafted using an inverse design methodology, a process that leverages
the characteristics of a hypersonic flow environment. This approach utilizes the
streamlines of the inviscid flow field post-shock as the basis for designing the compression
stream surface, typically the lower surface of the waverider geometries. The end result is
a streamlined geometry optimized for hypersonic flows. The geometry of the lower
surface, defined by a set of adjacent streamlines, ensures the leading edge remains
attached to the shock wave. Below is an outline of the algorithm used to generate
waverider geometries:
• Determining and designing the initial flow field in the direction of the flow.
• Solving for the flow field using computational fluid dynamics methods, such as the
Taylor–Maccoll equation [32], or the Euler, or Navier–Stokes equations.
• Mapping out the streamlines.
• Applying osculating cone theory along the spanwise direction.
Computation 2024, 12, 140 3 of 24

• Refining and finalizing the model design.

To specify the initial flow conditions from which the waverider is derived, a cone
with a predetermined angle 𝛿 is first designed. Then, the inlet boundary conditions like
Mach number, pressure, and temperature are set, and the resultant flow field is computed.
After calculating the flow field, the streamlines can be computed by numerically solving
the following set of ordinary differential equations (ODEs):
𝑑𝑥
= u(x,y,z)
𝑑𝑡
𝑑𝑦
= v(x,y,z) (1)
𝑑𝑡
𝑑𝑧
= w(x,y,z)
𝑑𝑡
Using the Runge–Kutta method, the above set of equations can be integrated in time
to obtain each streamline coordinate.
A schematic overview of the streamlines traced from a conical shock is depicted in
Figure 1. To establish a waverider geometry using the cone-derived method, one must
first define a base curve. This curve is then projected onto the shock, aligned parallel to
the freestream, to form the leading edge. Subsequently, the streamlines are traced
backward from the leading-edge positions on the shock to the base plane, creating the
waverider’s compression (lower) surface. Beyond the analytical Equation (1), an
alternative, simpler method can be employed to trace the streamlines by following
tangential paths to the base cone surface, thus facilitating the generation of the
compression surface geometry. The base curve equation, essential for designing the
waverider, is illustrated around the symmetry plane in Figure 2. The total width of the
waverider is denoted by 𝑙, with 𝑙 representing the flat section of the curve. The curved
section of the curve is determined by the equation depicted in Figure 2. The value of 𝑏 is
calculable from 𝑙, 𝑙 , and ℎ, values that are defined following the resolution of the initial
conical flow field.

Figure 1. Schematic representation of the waverider design methodology.

For a practical waverider design, addressing the sharpness of the leading edge is
crucial. Thus, a pragmatic design strategy incorporates blunting techniques to manage
this characteristic. Blunting the leading edge and the junctions where two stream surfaces
converge mitigates heating effects and results in a geometry more feasible for
manufacturing. However, blunting does deviate from the ideal waverider design,
Computation 2024, 12, 140 4 of 24

allowing some pressure leakage from the lower to the upper surface, which in turn
increases drag and diminishes the aerodynamic efficiency of the waverider. Studies [33,34]
have demonstrated that while blunting reduces the heat flux experienced by the vehicle,
it adversely impacts the lift–drag ratio (L/D). Despite this, a blunted waverider design can
still offer superior aerodynamic performance compared to other design alternatives. The
extent and nature of the blunting are crucial and require a balance between reducing
thermal stresses and optimizing aerodynamic efficiency. With various possible
configurations for a waverider, there are numerous approaches to designing a blunt
leading edge.
There are two primary methods for blunting the leading edge of a waverider: one
involves removing material from the edge, while the other entails adding material to it
[35]. Tincher and Burnett have proposed that the addition of material to the leading edge
incurs a lower aerodynamic penalty compared to material removal [35]. This approach is
favored because it allows for maintaining more of the original aerodynamic profile,
potentially minimizing the negative impacts on performance that are associated with
leading edge modifications.

Figure 2. Schematic representation of the waverider design methodology.

3. Direct Simulation Monte Carlo (DSMC) Methodology

The Direct Simulation Monte Carlo (DSMC) method, initially developed by Graeme
Bird in the late 1960s [36], employs a particle-based statistical approach to study rarefied
gas flows. Unlike traditional methods that solve equations to predict flow behavior,
DSMC simulates the motion and interactions of a large number of simulated particles,
each representing a fraction of the actual molecules in the flow field. This method
decouples the motion of the particles and computes intermolecular collisions over brief
time intervals. It has been mathematically demonstrated that with a sufficient number of
particles per cell, DSMC converges to the Boltzmann equation [37]. The DSMC technique
comprises four main steps:
1. Advancing the position of the particles.
2. Sorting the particles into respective cells.
3. Calculating the intermolecular collisions.
4. Gathering and analyzing the particle properties.
In practice, the computational domain is segmented into numerous cells, potentially
further divided into sub-cells based on a predefined grid. Initially, the particles are
Computation 2024, 12, 140 5 of 24

randomly assigned within each cell, with specific initial positions, velocities, and energy
states. These steps are cyclically repeated for each timestep to model the time evolution of
the flow. Macroscopic flow properties like velocity, pressure, density, and temperature
are derived from the weighted averages of these microscopic particle properties. For all
the flow simulations discussed in this study, the DSMC method as detailed by Graeme
Bird [36] and as implemented in the Stochastic Parallel Rarefied-gas Time-accurate
Analyzer (SPARTA) at Sandia National Laboratories [38], was utilized to examine the flow
characteristics.

4. Application to a Specific Waverider Design

In the development of a waverider designed for Mach 7, the initial step involved
crafting a 7-degree cone, 2 m in length with a 1 cm blunt leading edge, as depicted in
Figure 3. The three-dimensional flow field around this cone was then resolved using the
Direct Simulation Monte Carlo (DSMC) method. The simulation conditions were derived
from the US standard atmosphere model, representing the flow characteristics at an
altitude of 90 km. The specific flow conditions and the parameters used in the DSMC
simulation are detailed in Tables 1 and 2, respectively. In the present simulations, the gas
comprises 79% N2 and 21% O2; rotational and vibrational relaxation numbers are treated
as constants. The rotational relaxation number was set to Zrot = 5 for both N2 and O2,
whereas a vibrational relaxation number Zvib = 52,600 for N2 and Zvib = 17,900 for O2 have
been used. Elastic collisions are performed using the Variable Soft Sphere (VSS) model
[39] while inelastic collisions are performed using the Borgnakke–Larsen model [40]. For
the selection of collision partners, we utilized the near-collision partner algorithm. This
algorithm selects collision partners within a sphere whose radius is the distance a particle
travels in one timestep. This approach enables the use of a grid cell size larger than one
mean free path while maintaining collision accuracy [41] although our cell size is close to
1/3 of the mean free path. Due to the low temperatures occurring in these flight conditions,
chemical reactions are not enabled; thus, the flow is assumed to be in thermal equilibrium.
All the simulations were conducted in the ARCHER2 UK supercomputing facility.

Figure 3. The geometry of the 7-degree half cone used for the calculation of the initial flow field.

As it was previously mentioned, all the simulations were carried out using the open-
source parallel DSMC code SPARTA (Stochastic Parallel Rarefied-gas Time-accurate
Computation 2024, 12, 140 6 of 24

Analyzer), which was developed at Sandia National Laboratories [38]. After computing
the three-dimensional flow field around the 7-degree cone, the flow streamlines were
extracted at specific parallel planes to delineate the shockwave boundaries. Figure 4 offers
a visualization of these calculated flow field streamlines. The shock boundaries are
discernible by the bending of the flow streamlines in both the side and top planes, and by
a change in the color of the streamwise velocity flow field at the rear plane behind the
cone.

Figure 4. Streamlines of the three-dimensional flow field around the 7-degree cone. Side view (top),
top view (middle), rear view (bottom).
Computation 2024, 12, 140 7 of 24

Table 1. Parameters for a Mach 7 flow at 90 km altitude.

Mach number, Ma∞ 7

Reynolds, Re 2873
Freestream velocity, U∞ (m/s) 1965
Knudsen number, Kn 0.6
Angle of attack, α (degrees) 0
Freestream density, ρ∞ (kg/m3) 9.62 × 10-6
Freestream pressure, p∞ (Pa) 0.55
Freestream temperature, T∞ (K) 196
Wall temperature, Tw (K) 300
Gas Air

Table 2. DSMC computational parameters for Mach 7 flow at 90 km altitude.

Number density, Nd∞ (#/m3) 2.00 × 1020

Timestep (s) 3.0 × 10-7
Transient period 40,000
Sample period 100,000
Number of particles 10.5 × 109
Number of cells 183,664,000
Wall-clock time (CPU hours) 195,840

After obtaining the flow planes from the DSMC simulation, these are imported into
a Computer-Aided Design (CAD) software (SolidWorks 2021) to precisely define the
geometry of the waverider. The design process begins at the rear of the vehicle, crafting
curves that are tangent to the observed circular shock shape, with the first characteristic
cross-section displayed in Figure 5 (bottom) and Figure 6. As seen in Table 1, due to the
low Reynolds number the flow falls within the laminar regime.
To ensure the design remains practical for real-world applications, the leading edges
of the waverider are fashioned to be blunt rather than sharp. To accurately control the
bluntness of the leading edges, a four-point Bezier curve is utilized to sculpt the leading-
edge profile. This method provides the flexibility to finely adjust the curvature and thus
tailor the aerodynamic and thermal properties of the edge. Similarly, additional sections
along the waverider’s longitudinal axis are designed at intervals of every 30 cm. Both the
upper and lower surfaces of the waverider are defined by the shockwave boundaries
generated from the initially prescribed 7-degree cone geometry. This approach ensures
that the final waverider design is optimized within the constraints set by the initial flow
conditions and the aerodynamic needs of hypersonic travel.
Computation 2024, 12, 140 8 of 24

Figure 5. Waverider surface along the flow streamlines, in comparison with the initial cone. Top
view (top), side view (middle), and rear view (bottom).

Figure 6. Back section of the waverider (units in mm).

Figure 7 provides a visual representation of the various sections of the waverider

geometry, although not all the sections are displayed to maintain clarity. Upon completing
Computation 2024, 12, 140 9 of 24

the design of all the sections, a loft operation with tangency constraints is employed to
generate the surfaces of the waverider. To simplify the design process and leverage the
symmetry of the flow field, only half of the final geometry is initially designed. This half
is then mirrored around the symmetry plane to produce the complete vehicle structure.

Figure 7. Surface lofts along the vehicle profiles (top). Vehicle overview without the nose section
(bottom).

The methodology of the osculating cones, as described earlier, serves as a robust

computational framework for designing hypersonic waveriders. Hypersonic
aerodynamics involves dealing with extreme temperatures, high pressures, and intricate
shock wave interactions, making the design process inherently complex. By breaking
down the methodology into discrete steps, this approach allows for a detailed focus on
both the theoretical and practical aspects, guiding the creation of these advanced
hypersonic vehicles. This structured design process ensures that each aspect of the
vehicle’s performance and structural integrity is considered, leading to the development
of effective and efficient hypersonic waveriders.
The design process for a hypersonic waverider involves a meticulous and structured
approach, crucial for achieving aerodynamic efficiency and structural integrity under
extreme conditions. Here is a detailed breakdown of each step in the design procedure:
Step 1: Initiating at the Aft End of the Cone
The design process begins at the trailing end of a cone, which is pivotal for setting
the initial conditions of the flow field. Here, the shock wave boundaries are meticulously
defined, shown as dashed white lines in Figure 4 (bottom). This foundational step is
critical as it establishes the first profile of the waverider. The design ensures that both the
leading edge and the lower surface of this profile are tangentially aligned with the shock
wave boundaries, minimizing drag and enhancing aerodynamic efficiency.
Step 2: Sequential Cross-section Profiling
Computation 2024, 12, 140 10 of 24

Moving forward, the cross-sectional planes are taken at intervals of about 30 to 40 cm

along the cone’s length. Each plane provides a template for shaping a segment of the
vehicle’s body, tailored to the specific aerodynamic conditions of that segment. This step
is akin to sculpting the individual slices of the vehicle, each optimized for its particular
flow dynamics.
Step 3: Lofting Operation
The individual profiles generated in Step 2 are then interconnected through a lofting
operation. This technique merges the profiles to form the continuous outer skin of the
waverider, as seen in Figure 7. The operation uses interpolation methods to ensure a
smooth contour transition between sections. However, the profiles narrow as they
approach the cone’s apex, which presents challenges in maintaining the structural and
aerodynamic continuity.
Step 4: Addressing the Nose Region
In the nose region, where profiles become exceedingly slender, traditional lofting
techniques are less effective. To address this, boundary surfaces that conform to the
curvature of the original cone are introduced. As depicted in Figure 8, these surfaces form
the nose section’s outer skin, integrating seamlessly with the lofted sections to maintain
aerodynamic integrity.
Step 5: Knitting Surfaces for Fluid Simulation
Following the design of the nose and main body sections, a knitting operation is
employed to merge all the surface segments into a unified, watertight structure. This
integration is crucial for the accuracy of subsequent fluid dynamics simulations, ensuring
no gaps or inconsistencies exist that could compromise the model’s stability or precision.
The completion of this step results in a complete vehicle geometry, illustrated in Figure 9.
Step 6: Mirroring Operation
To enhance design efficiency and reduce computational demands, only half of the
vehicle is modeled initially. The other half is created through a mirroring operation along
a symmetry plane. This strategy not only simplifies the construction process but also
facilitates faster iterations and refinements during the design phase, effectively halving
the computational resources required.
Step 7: Refined Leading-Edge Design
This step in the waverider design methodology underscores the importance of
optimizing the leading edges, which play a critical role in the vehicle’s overall
performance and survivability in hypersonic conditions. Unlike the typical hypersonic
designs that feature sharp leading edges to minimize drag and maximize speed, this
approach recognizes the trade-offs such sharp edges bring, including high stress
concentrations and reduced thermal resilience. These factors can lead to structural failures
when subjected to the intense heat and pressure of hypersonic flight.
Computation 2024, 12, 140 11 of 24

Figure 8. Nose section with upper boundary surface.

To address these challenges, the design incorporates three-point Bezier curves to

blunt the leading edges. This technique allows for precise control over the curvature and
tangency of the edges, making it possible to tailor them specifically to the demands of
high-speed aerodynamics and thermal stress management. The Bezier curves not only
help in smoothing the transition of the flow over the vehicle but also enhance the thermal
robustness by distributing the stresses and heat more evenly across the surface. The use
of Bezier curves in this context provides a balance between aerodynamic efficiency and
structural integrity, ensuring the leading edges are not only effective in minimizing
aerodynamic drag but are also capable of withstanding the harsh conditions of hypersonic
flight. This nuanced approach results in a leading-edge design that is both geometrically
optimized and thermodynamically durable, setting a new standard for hypersonic vehicle
design. Through these steps, the waverider design is developed with precision,
addressing both the high demands of hypersonic travel and the practical aspects of vehicle
manufacture and simulation.
The osculating cones methodology is a sophisticated framework that draws from the
fields of aerodynamics, computational fluid dynamics (CFD), and applied mathematics.
This approach forms a holistic and in-depth pathway for designing hypersonic
waveriders, effectively addressing the complex challenges associated with hypersonic
flows and structural dynamics. The integration of these multi-disciplinary domains
ensures that the resulting waverider design achieves not only aerodynamic efficiency but
also exhibits enhanced thermal and structural robustness.
Computation 2024, 12, 140 12 of 24

Figure 9. (Top): Waverider sections and the complete geometry. (Bottom): Flowchart of the design
methodology.

In the first phase of this computational exploration, the creation of an accurate

computational mesh is crucial. This mesh is instrumental in precisely capturing the
intricate geometric details of the waverider, particularly in the areas of high sensitivity,
such as the leading edges. The precision in modeling these edges is paramount because
even slight inaccuracies can lead to significant aerodynamic repercussions. This
meticulousness in the mesh design is essential for replicating the intended geometrical
features and ensuring that the simulations reflect the true aerodynamic behavior of the
waverider. The computational surface mesh used in the Direct Simulation Monte Carlo
(DSMC) method is specifically tailored to accommodate the unique characteristics of the
waverider’s design, as illustrated in Figure 10. This mesh allows for the detailed and
accurate simulation of the flow dynamics around the waverider, capturing the essential
interactions between the vehicle’s surface and the hypersonic flow. The DSMC method,
known for its ability to model rarefied gas dynamics effectively, is particularly well-suited
for analyzing the aerodynamic properties of vehicles operating at the upper edges of the
atmosphere, where conventional CFD methods might struggle. For the simulation of the
final waverider design, we used the SPARTA DSMC code, consistent with the simulation
of the initial shock-generating cone. This consistency was maintained because the flight
Computation 2024, 12, 140 13 of 24

conditions did not change. As shown in Figure 11 (bottom) the Knudsen number falls
within the upper end of the slip regime or within the transition regime.

Figure 10. The utilized surface mesh. Lower surface (top) and isometric view (bottom).
Computation 2024, 12, 140 14 of 24

Figure 11. Pressure contours around the vehicle (top), Knudsen number of the flow field based on
the waverider length (bottom).

5. Simulation Results
Due to the extensive computational demands associated with simulating hypersonic
flows, several strategies are implemented to optimize the use of computational resources.
One effective technique is the simulation of only half of the waverider’s geometry, with
the resulting flow field then being symmetrically mirrored across a predefined plane. This
approach significantly reduces the computational load without sacrificing the accuracy of
the results. A key objective of these simulations is to confirm the correct positioning of the
shock wave relative to the vehicle’s leading edges. This verification is conducted by
analyzing the pressure contours and Q-criterion plots, which are displayed in Figure 11
and Figure 12, respectively.
The Q-criterion, calculated using Equation (2), involves Ω the antisymmetric
component of the velocity gradient representing vorticity, and 𝑆 , the symmetric
component representing the rate of strain.

𝑄 = (‖𝛺‖ − ‖𝑆‖ ). (2)

According to the established criteria, positive Q-criterion values signify regions

within the flow field where vorticity predominates, suggesting the areas of rotational
flow. Conversely, negative values indicate the dominance of strain rate or viscous stresses
[42,43]. This criterion is a critical diagnostic tool for analyzing the flow characteristics
around the waverider. For instance, in Figure 13, the negative Q-criterion values observed
below the leading edge signal the presence of a shock in that area. The associated pressure
contour plots, illustrated in Figure 11, reveal a distinct pressure buildup around the
leading edges of the waverider, confirming the effective attachment of the shock wave to
the vehicle’s geometry.
Computation 2024, 12, 140 15 of 24

Figure 12. Q-criterion contours around the vehicle.

Moreover, Figure 11 graphically depicts the pressure distribution on the surface of

the waverider. As anticipated, increased pressure levels are prominent around the leading
edges and nose section, validating the presence of an attached shock wave. This
integration of various computational diagnostics—from mesh optimization to detailed
pressure and Q-criterion analyses—offers a thorough assessment of the waverider’s
aerodynamic performance. These results underscore the computational rigor and physical
validity of the waverider design, particularly regarding the attachment of shock waves to
the leading edges. Figure 13 depicts the Q-criterion contours on the plane of symmetry,
while Figure 14 shows the corresponding contours on a vertical plane at the rear of the
vehicle, where the concentrated blue regions at the wing tips and under the fuselage
indicate attached shocks, further demonstrating the aerodynamic efficacy of the design.

Figure 13. Q-criterion contours on the plane of symmetry.

Computation 2024, 12, 140 16 of 24

Figure 14. Q-criterion contours on a vertical plane.

Figures 15 and 16 present the streamwise velocity contours and the temperature field
around the vehicle. These figures are crucial for understanding how the vehicle interacts
with the flow at high speeds. In particular, the simulations reveal important thermal
dynamics and fluid flow characteristics. For the specific flow conditions applied in this
study, the temperature field around the vehicle shows significant thermal variations. A
peak temperature of 1400 Kelvin is observed just in front of the nose section of the
waverider, indicating areas where aerodynamic heating is most intense due to the rapid
flow deceleration. This information is vital for assessing the thermal resistance and
structural integrity of the waverider under operational conditions.

Figure 15. Streamwise velocity contours on the symmetry plane of the waverider.
Computation 2024, 12, 140 17 of 24

Figure 16. (Top): total temperature field around the vehicle. (Bottom): rotational temperature field
(on the symmetry plane of the waverider).

In Figures 17–20, various flow properties are illustrated on a plane parallel to the
waverider, with the surface of the vehicle distinctly highlighted by the variations in the
surface pressure. Specifically, Figure 17 displays the pressure contours on this plane,
providing a clear visualization of how the pressure varies in response to the shape and
aerodynamics of the vehicle. Notably, considering that the free-stream pressure is
established at 0.55 Pa, there is a significant pressure increase observed around the nose of
the waverider, rising to an order of magnitude higher than the ambient conditions.
Additionally, the pressure around the leading edges of the waverider is approximately
four times higher than that of the free-stream pressure. This detailed mapping of pressure
distribution is critical for evaluating the aerodynamic performance and structural stresses
experienced by the waverider during operation.
Computation 2024, 12, 140 18 of 24

Figure 17. Pressure contours on a horizontal plane parallel to the vehicle.

Figure 18 contains the velocity contours on the same plane. In this figure, the
attachment of the shock to the leading edge and the nose of the vehicle can be clearly
observed. As can be observed in Figure 19, the temperature around the leading edges is
800 K, which is about four times higher than the free-stream temperature of 196 K (with a
peak of around 1400 K at the nose of the vehicle).

Figure 18. Streamwise velocity contours on a horizontal plane parallel to the vehicle.

Figure 19. Temperature contours on a horizontal plane parallel to the vehicle.

Computation 2024, 12, 140 19 of 24

In Figure 20, the Q-criterion is meticulously illustrated on the same plane parallel to
the waverider, effectively capturing the entire field of the attached shock wave. This
visualization underscores the precise areas where the shock remains attached to the
surface of the waverider, a critical aspect of its aerodynamic design. Additionally, the
positive values of the Q-criterion in this figure indicate the formation of two distinct
vortices, located symmetrically, one on each side, at the tip of the vehicle’s rear section.
These vortices are pivotal in understanding the flow dynamics at the rear of the waverider,
particularly how they influence the overall stability and control of the vehicle at high
speeds. For a more detailed observation of these vortices, Figure 21 provides an enhanced
view where the vorticity at the rear of the waverider is plotted. In this figure, the two
vortices are distinctly highlighted in red, offering a stark contrast against the rest of the
flow field. This color differentiation not only emphasizes the location and intensity of the
vortices but also aids in analyzing their structure and interaction with the surrounding
flow. The depiction of these vortices in such vivid detail is crucial for engineers and
designers to evaluate the aerodynamic interactions at play and to refine the waverider’s
design for optimized performance.

Figure 20. Q-criterion contours on a horizontal plane parallel to the vehicle.

Figure 21. Vorticity magnitude at the back of the vehicle.

Given the intricate design of the waverider, it was deemed necessary to employ the
three-dimensional iso-surfaces of the Q-criterion to fully capture the complex spatial
Computation 2024, 12, 140 20 of 24

characteristics of the shocks and the structured vortices surrounding the vehicle. Figure 22
illustrates these three-dimensional Q-criterion contours enveloping the vehicle, with the
contours colored by vorticity magnitude to emphasize the variations in the rotational flow
behavior. The transparent grey iso-surfaces distinctly outline the external shape of the shock
waves, providing a comprehensive view of their three-dimensional form. Notably, parts of
the shock wave close to the compression surface are also visible in the same figure, offering
a detailed look at how the shock interacts with the vehicle’s structure.
In Figure 23, these contours are similarly plotted but are instead colored by velocity
magnitude to highlight the differences in flow speed. As depicted in this figure,
approximately 30 cm downstream from the nose of the vehicle—where the compression
surface becomes steeper—the vortices detach from the main flow and continue
downstream towards the higher pressure regions at the back of the vehicle. This
separation is critical for understanding the aerodynamic behavior of the waverider.
Additionally, as the angle of the compression surface increases towards the vehicle’s rear,
a noticeable pressure rise occurs, and the flow velocity significantly reduces, slowing
down by about 1000 m/s within just a few centimeters. These detailed visualizations and
analyses of flow dynamics are essential for assessing the performance implications of the
vehicle’s design under operational conditions.

Figure 22. Three-dimensional Q-criterion contours, colored by vorticity magnitude.

Figure 23. Overview of the three-dimensional Q-criterion contours around the vehicle, colored by
velocity magnitude.
Computation 2024, 12, 140 21 of 24

To determine the aerodynamic forces acting on the waverider, the shear stresses and
pressure components across its three-dimensional surface were meticulously calculated.
These calculations facilitate the precise measurement of lift and drag forces per unit
surface area, with the results displayed in Figure 24. The analysis of this figure reveals
that the maximum drag primarily occurs at the tip of the nose, while a significant amount
of lift is generated in the area just below the nose. By integrating the forces measured
across all the surface cells, the total lift and drag for the waverider were computed,
yielding a lift–drag ratio (L/D) of 2.18. According to existing studies, this L/D ratio is
typical for a waverider operating at a 0-degree angle of attack [44], indicating that a value
above 2.0 is indicative of effective aerodynamic performance under the chosen design
conditions.
The Knudsen number contours, based on the waverider length, are depicted in Figure
11 (bottom), while the Mach number contours at the plane of symmetry are depicted in
Figure 25. As demonstrated the Knudsen number around the vehicle is within the
transitional regime, which justifies the use of the DSMC method.
For this study, only a single angle of attack was investigated; however, the authors in
[44] suggest that a slight increase in the angle of attack could potentially enhance the L/D
ratio by two to three times, thereby significantly improving performance. It is also
noteworthy that waveriders are commonly analyzed at altitudes between 25 km and 40
km in the open literature, unlike the 90 km altitude considered in this study. The elevated
altitude impacts the generation of lift due to different fluid dynamics, necessitating a
larger compression surface than what is typically seen in waveriders operating at lower
altitudes. This adaptation is a critical design modification that enhances the vehicle’s
performance in the rarified atmospheric conditions encountered at high altitudes.

Figure 24. Overview of the lift (bottom) and drag (top) per unit surface, exerted on the waverider’s
lower surface.
Computation 2024, 12, 140 22 of 24

Figure 25. Mach number contours around the vehicle at the plane of symmetry.

6. Conclusions
In the comprehensive design and analysis of a Mach 7 waverider, intricate
computational methodologies were employed to accurately simulate the vehicle’s
aerodynamic properties. After applying the osculating cone methodology, for the
definition of the waverider’s geometry, a precise computational mesh was constructed to
capture the complex surface geometry, particularly around the waverider’s critical
leading edges. An advanced simulation technique, the Direct Simulation Monte Carlo
(DSMC) method, was used to resolve the three-dimensional flow field around the vehicle.
This simulation not only detailed the pressure and thermal conditions surrounding the
waverider but also highlighted the structural interactions of the shock waves, validated
through the visualizations of streamwise velocity and temperature fields.
Further refinement of the simulation model involved creating the three-dimensional
iso-surfaces of the Q-criterion to intricately map out the shock and vortex structures
around the waverider. This allowed for an in-depth examination of the flow dynamics,
particularly the attachment of shock waves and the formation of vortices at the critical
sections of the vehicle. Such detailed analyses were essential for confirming the
aerodynamic efficiency of the design. Additionally, the use of layered meshing techniques
facilitated the accurate capture of boundary layer phenomena, crucial for assessing the
vehicle’s performance under the extreme conditions of hypersonic flight.
The culmination of these extensive simulations was the calculation of lift and drag
forces, where the lift–drag ratio was determined to be 2.18, indicative of efficient
aerodynamic performance for a waverider at a zero-degree angle of attack. These findings
are significant, as they not only demonstrate the waverider’s capability under the
stipulated design conditions but also suggest potential improvements for future designs.
For instance, adjustments in the angle of attack could further optimize the vehicle’s
performance. This study’s insights into high-altitude aerodynamics, where different fluid
dynamics necessitate modifications such as larger compression surfaces, pave the way for
the enhanced designs of future hypersonic vehicles operating in near-space environments.

Author Contributions: Conceptualization, A.K. and I.K.N.; methodology, A.K. and I.K.N.; software,
A.K.; validation, A.K. and I.K.N.; resources, A.K.; writing—original draft preparation, A.K.;
writing—review and editing, I.K.N.; visualization, A.K.; supervision, I.K.N.; project administration,
I.K.N.; funding acquisition, I.K.N. All authors have read and agreed to the published version of the
manuscript.
Computation 2024, 12, 140 23 of 24

Funding: This study was funded by the Integrated Air and Missile Defence Centre of Excellence
(IAMD COE) and any intellectual property resulting from the work covered by this will be property
of the IAMD COE. This paper reflects only the IAMD COE policies and its author(s)’ positions and
it is not intended to create any legal obligations nor does it reflect NATO’s policies or positions, or
engage NATO in any way.
Data Availability Statement: The data presented in this study are available on request from the
corresponding author due to the fact that this project was funded by NATO IAMD-COE, and
sharing the data requires their explicit permission.
Acknowledgments: The authors would like to thank the UKTC and EPSRC for computational time
on the UK supercomputing facility ARCHER2 via projects EP/R029326/1 and EP/X035484/1.
Conflicts of Interest: The authors declare no conflicts of interest.

References
1. Sandlin, D.R.; Pessin, D.N. Aerodynamic Analysis of Hypersonic Waverider Aircraft; NASA-CR-192981; NASA: Washington, DC,
USA, 1993.
2. Nonweiler, T.R.F. Delta wings of shape amenable to exact shock wave theory. J. R. Aero. Soc. 1963, 67, 39.
3. Sziroczak, D.; Smith, H. A review of design issues specific to hypersonic flight vehicles. Prog. Aerosp. Sci. 2016, 84, 1–28.
4. Zhang, T.T.; Wang, Z.G.; Huang, W.; Yan, L. Parameterization and optimization of hypersonic-gliding vehicle configurations
during conceptual design. Aerosp. Sci. Technol. 2016, 58, 225–234.
5. Kontogiannis, K. On Developing Efficient Parametric Geometry Models for Waverider-Based Hypersonic Aircraft. Ph.D. Thesis,
University of Southampton, Southampton, UK, 2017.
6. Son, J.; Son, C.; Yee, K. A Novel Direct Optimization Framework for Hypersonic Waverider Inverse Design Methods. Aerospace
2022, 9, 348.
7. Flower, J.W. Configurations for high supersonic speeds derived from simple shockwaves and expansions. R. Aeronaut. Soc. J.
1963, 67, 287–290.
8. Kuchemann, D. The Aerodynamic Design of Aircraft: A Detailed Introduction to the Current Aerodynamic Knowledge and Practical Guide
to the Solution of Aircraft Design Problems; Pergamon Press: Oxford, UK, 1978.
9. Bowcutt, K.G. Multidisciplinary optimization of airbreathing hypersonic vehicles. J. Propuls. Power 2001, 17, 1184–1190.
10. Cockrell, C.E., Jr. Interpretation of waverider performance data using computational fluid dynamics. J. Aircr. 1994, 31, 1095–
1100.
11. Corda, S.; Anderson, J., Jr. Viscous optimized hypersonic waveriders designed from axisymmetric flow fields. In Proceedings
of the 26th Aerospace Sciences Meeting, Reno, NV, USA, 14 January 1988.
12. Nonweiler, T.R.F. Aerodynamic problems of manned space vehicles. R. Aeronaut. Soc. J. 1959, 63, 521–528.
13. Zubin, M.A.; Ostapenko, N.A. Experimental investigation of some singularities of the supersonic flow around V-shaped wings.
Fluid Dyn. 1975, 10, 647–652.
14. Ostapenko, N.A. Aerodynamic characteristics of V-shaped wings with shock waves detached from leading edges at hypersonic
speeds. Fluid Dyn. 1993, 28, 545–552.
15. Maksimov, F.A.; Ostapenko, N.A. V-shaped wings with an opening angle exceeding π at super- and hypersonic flow velocities.
Dokl. Phys. 2016, 61, 412–417.
16. Mazhul, I.I.; Rakhimov, R.D. Numerical investigation of off-design regimes of flow past power-law waveriders based on the
flows behind plane shocks. Fluid Dyn. 2003, 38, 806–814.
17. Mazhul, I.I.; Rakhimov, R.D. Hypersonic power-law shaped waveriders. J. Aircr. 2004, 41, 839–845.
18. Mazhul, I.I. Off-design regimes of flow past waveriders based on isentropic compression flows. Fluid Dyn. 2003, 45, 271–280.
19. Jones, J.G.; Moore, K.C.; Pike, J.; Roe, P.L. A method for designing lifting configurations for high supersonic speeds, using
axisymmetric flow fields. Ing.-Arch. 1968, 37, 56–72.
20. Mangin, B.; Benay, R.; Chanetz, B.; Chpoun, A. Optimization of viscous waveriders derived from axisymmetric power-law blunt
body flows. J. Spacecr. Rocket. 2006, 43, 990–998.
21. Corda, S. Viscous Optimized Hypersonic Waveriders Designed from Flows over Cones and Minimum Drag Bodies. Ph.D.
Thesis, University of Maryland, College Park, MD, USA, 1988.
22. Ding, F.; Liu, J.; Shen, C.-B.; Huang, W. Novel approach for design of a waverider vehicle generated from axisymmetric
supersonic flows past a pointed von Karman ogive. Aerosp. Sci. Technol. 2015, 42, 297–308.
23. Rasmussen, M.L. Lifting-body Configurations Derived from Supersonic Flows Past Inclined Circular and Elliptic Cones. In
Proceedings of the 5th Atmospheric Flight Mechanics Conference for Future Space Systems, AIAA Paper, Boulder, CO, USA,
6–8 August 1979; p. 1665.
24. Rasmussen, M.L. Waverider configurations derived from inclined circular and elliptic cones. J. Spacecr. Rocket. 1980, 17, 537–545.
25. Rasmussen, M.L.; Jischke, M.C.; Daniel, D.C. Experimental forces and moments on cone-derived waveriders for m = 3 to 5. J.
Spacecr. Rocket. 1982, 19, 592–598.
Computation 2024, 12, 140 24 of 24

26. Rasmussen, M.L.; Jischke, M.C.; Daniel, D.C. Experimental surface pressures on cone-derived waveriders for M = 3–5. J. Spacecr.
Rocket. 1983, 20, 539–545.
27. Lin, S.C.; Rasmussen, M.L. Cone-derived Waveriders with Combined Transverse and Longitudinal Curvature. In Proceedings
of the 26th Aerospace Sciences Meeting, AIAA Paper 88–03714, Reno, NV, USA, 11–14 January 1988.
28. Liu, C.Z.; Bai, P.; Chen, B.Y.; Ji, C.Q. Rapid Design and Optimization of Waverider from 3D Flow. In Proceedings of the 16th
AIAA Aviation Technology, Integration, and Operations Conference, AIAA Paper 2016–3288, Washington, DC, USA, 13–17 June
2016.
29. Mclaughlin, T.A. Viscous Optimized Hypersonic Waveriders for Chemical Equilibrium Flow. Master’s Thesis, University of
Maryland, College Park, MD, USA, 1990.
30. Anderson, J.D.; Chang, J.; Mclaughlin, T.A. Hypersonic Waveriders: Effects of Chemically Reacting Flow and Viscous
Interaction. In Proceedings of the 30th Aerospace Sciences Meeting and Exhibit, AIAA Paper 92–0302, Reno, NV, USA, 6–9
January 1992.
31. Li, M.Y.; Zhou, H. Research on the design method of waverider aerodyanmic configuration based on commercial CFD software.
Chin. Q. Mech. 2014, 35, 293–299.
32. Taylor, G.I.; Macoll, J.W. The air pressure on a cone moving at high speeds I. Proc. R. Soc. A 1933, 139, 278–297.
33. Santos, W.F. Bluntness impact on lift-to-drag ratio of hypersonic wedge flow. J. Spacecr. Rocket. 2009, 46, 329–339.
34. Chen, X.Q.; Hou, Z.X.; Liu, J.X.; Gao, X.Z. Bluntness impact on performance of waverider. Comput. Fluids 2011, 48, 30–43.
35. Tincher, D.J.; Burnett, D.W. Hypersonic waverider test vehicle-a logical next step. J. Spacecr. Rocket. 1994, 31, 392–399.
36. Bird, G.A. Molecular Gas Dynamics and the Direct Simulation of Gas Flows, Clarendon Press: Oxford, UK, 1994.
37. Wagner, W. A convergence proof for Bird’s direct simulation Monte Carlo method for the Boltzmann equation. J. Stat. Phys.
1992, 66, 1011–1044.
38. Gallis, M.A.; Torczynski, J.R.; Plimpton, S.J.; Rader, D.J.; Koehler, T. Direct Simulation Monte Carlo: The quest for speed. In
Proceedings of the 29th Rarefied Gas Dynamics (RGD) Symposium, Xi’an, China, 13–18 July 2014.
39. Koura, K.; Matsumoto, H. Variable soft sphere molecular model for air species. Phys. Fluids A Fluid. Dyn. 1992, 4, 1083–1085.
40. Borgnakke, C.; Larsen, P.S. Statistical Collision Model for MonteCarlo Simulation of Polyatomic Gas Mixture. J. Comput. Phys.
1975, 18, 405–420.
41. Gallis, M.A.; Torczynski, J.R. Effect of Collision-Partner Selection Schemes on the Accuracy and Efficiency of the Direct
Simulation Monte Carlo Method. Int. J. Numer. Methods Fluids 2011, 67, 1057–1072. https://doi.org/10.1002/fld.2409.
42. Zhan, J.M.; Li, Y.T.; Wai, W.H.O.; Hu, W.Q. Comparison between the Q criterion and Rortex in the application of an in-stream
structure. Phys. Fluids 2019, 31, 121701.
43. Jeong, J.; Hussain, F. On the identification of a vortex. J. Fluid Mech. 1995, 285, 69–94.
44. Huang, W.; Ma, I.; Wang, Z.; Pourkashanian, D.B.; Luo, S.; Lei, J.A. parametric study on the aerodynamic characteristics of a
hypersonic waverider vehicle. Acta Aeronaut. 2011, 69, 135–140.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury
to people or property resulting from any ideas, methods, instructions or products referred to in the content.