International Journal of

Turbomachinery
Propulsion and Power

Article
Metamodel-Assisted Multidisciplinary Design
Optimization of a Radial Compressor †
Mohamed H. Aissa * and Tom Verstraete
Department of turbomachinery and Propulsion, Von Karman Institute for Fluid Dynamics,
1640 Sint-Genesius-Rode, Belgium; tom.verstraete@vki.ac.be
* Correspondence: aissa@vki.ac.be
† This paper is an extended version of our paper published in Proceedings of the European Turbomachinery
Conference ETC13, Lausanne, Switzerland, 8–12 April 2019, Paper No. 104.




Received: 17 September 2019; Accepted: 30 October 2019; Published: 3 November 2019


Abstract: Kriging is increasingly used in metamodel-assisted design optimization. For expensive

simulations; however, one can afford only a few samples to build the Kriging model, which consequently
lacks prediction accuracy. We propose a bounded Kriging able to handle optimization problems with
a small initial database. During the optimization, the proposed Kriging suggests designs close to
database samples and finds optimal designs while staying in a feasible region (with respect to mesh and
CFD convergence). The bounded Kriging is applied along with the ordinary Kriging to a multidisciplinary
design optimization of a radial compressor. The shape of the compressor blades is optimized by
considering the aero performance at different operating points and the mechanical stresses. The objective
of the optimization is to maximize the efficiency at two operating points, while constraints are imposed
on the maximum stress level in the material, the choke mass flow, the pressure ratio and the momentum
of inertia of the impeller. While ordinary Kriging stopped prematurely because of many failing design
evaluations, the bounded Kriging satisfied all constraints and reached an improvement of 2.59% in
efficiency over the baseline design that does not comply with any constraints. The bounded Kriging
covers a special need for robust methods in optimization able to deal with challenging geometries and a
small database, which is the case for most industrial design optimizations.

Keywords: multidisciplinary optimization; radial compressor; Kriging; metamodel

1. Introduction
Traditional approaches for the design of radial compressors start with an aerodynamic design, which
is afterward submitted to a finite element stress analysis to verify that the maximum allowable stress in
the material has not been exceeded. If so, the geometry is returned to the aero designer with restrictions
on the design space. Several iterations between the aero design and the mechanical analysis may be
needed before a satisfactory compromise is found. The concurrent design methodology, used in this work,
addresses both the aerodynamic objectives and the material constraints in a single shot. The methodology
builds on high-fidelity numerical evaluations coupled with strong optimization methods and reduces
considerably the total design time.
Optimization methods can be classified according to their use of gradient information into
gradient-based and gradient-free methods. Gradient-based methods are the fastest to converge with

Int. J. Turbomach. Propuls. Power 2019, 4, 35; doi:10.3390/ijtpp4040035 www.mdpi.com/journal/ijtpp

Int. J. Turbomach. Propuls. Power 2019, 4, 35 2 of 20

a risk of being trapped in a local optimum [1]. Gradient-free methods are known to be global but have a
very slow convergence rate not suitable for time-consuming simulations.
Among gradient-free methods, Evolutionary Algorithms (EA), developed in the late 1960s [2], have
certain advantages above gradient-based methods, when applied to design optimization problems.
They are noise-tolerant, do not require the objective function to be continuous and can converge to
the global optimum. EAs methods have, however, a slow convergence rate resulting in a large number of
function evaluations. To improve the convergence rate of evolutionary methods, it is possible to make
use of metamodels, which are cheap heuristic models not based on physics. Those models absorb the
accumulated knowledge of a design space by evaluating samples and afterward providing predictions
for non-evaluated areas of the design space. As the predictions are computationally cheap, it is possible
to thoroughly scan the surrogate objective space instead of performing time-consuming simulations.
However, metamodels are not suitable for extrapolation and provide non-accurate predictions for designs
far away from sampled areas in the design space. High-fidelity simulations are then used to provide the
metamodel with more data on unsampled areas in order to enhance its accuracy. Such techniques are
referred to as Metamodel Assisted Evolutionary Algorithms.
Within a Metamodel-Assisted EA (MAEA), the Evolutionary Algorithm (EA) does not manipulate
directly the designs by running high-fidelity evaluations. Its role is rather confined to scan the prediction
space for an optimal prediction (e.g., minimal losses and maximal efficiency). Metamodels could rely on
Artificial Neural Networks (ANN), radial basis function, Kriging and many other methods. Kriging, which is
an interpolation method initially used to estimate mineral concentrations based on scarce data [3], shows a
better accuracy when dealing with small datasets as the ones emerging from time-consuming simulations.
Due to its capability of not only predicting a value but also giving the uncertainty of the prediction,
the method became of interest to approximate deterministic computer models. Moreover, several authors
use Kriging methods to accelerate the optimization process of turbomachinery applications [4–6]. The method
has since then evolved in many variants such as ordinary Kriging, universal Kriging, co-Kriging, blind
Kriging and many others [7,8].
Metamodel-assisted optimization, especially with Kriging, has been used successfully in
many disciplines. The use of the metamodel improves the convergence of the optimization compared
to a pure gradient-free solver and is a must for time-consuming simulations, which is the case for most
realistic application such as in aerodynamics, turbomachinery and automotive. At the beginning of the
optimization, when few samples are available, e.g., from a Design of Experiment (DOE), the metamodel
provides low prediction accuracy because Kriging is less accurate away from resolved areas [9].
The prediction accuracy improves when more samples are evaluated. In some cases with complex
geometries, the proposed designs at the beginning of the optimization could be problematic to evaluate
with the high-fidelity simulation. Even if the parameterization were checked for geometrical constraints,
there would be no guarantee that a design space has no challenging designs unless a time consuming
procedure of design space scanning is undertaken. In practice, FEM and CFD simulations of such designs
could have convergence problems and some designs could be challenging even for the meshing. In case an
evaluation is not possible, no feedback reaches the metamodel which will then keep proposing the same
failing design in an infinite loop. When noticing the failed simulation, it is possible to push the search out
of the problematic region by an intrusive penalty strategy; however, this distorts the objective space and
may hide any optimum located next to the area of the failed simulation [10].
This article tackles this issue by introducing a bounded Kriging able to handle evaluations with
meshing or convergence issues. For the bounded Kriging, we limit the hyper-parameters intervals during
the maximum likelihood training in a way to confine Kriging to stay next to explored areas and move only
gradually away. As a result, the Kriging model will suggest designs not far from known samples and
finds optimal designs while staying in a feasible region (with respect to mesh and CFD convergence) for
Int. J. Turbomach. Propuls. Power 2019, 4, 35 3 of 20

most of the optimization. The bounded Kriging along with the ordinary Kriging have been applied to a
metamodel-assisted multidisciplinary design optimization of a highly constrained radial compressor. The
bounded Kriging guided first the MAEA optimization algorithm toward regions of feasible designs before
identifying better performing designs. The ordinary Kriging, however, did not reach satisfying results and
had to be stopped after proposing a series of designs that failed during the aerodynamic evaluation.
The rest of the article is structured as follows: first the Kriging metamodel is introduced, and then the
application of the radial compressor is presented. Within the Application Section, the definition of the
geometry, the FVM and FEM evaluation and objective of the optimization are discussed first, followed by
the results of the multidisciplinary optimization using the bounded Kriging.

2. Methodology
This section introduces first the metamodel-assisted optimization method with a focus on the database
and the model update. Then, Kriging is presented and some aspects are highlighted such as the exploration
versus exploitation approach and the way failed simulations are treated.

2.1. Metamodel-Assisted Evolutionary Algorithm (MAEA)

Algorithm 1 shows the MAEA optimization method, which contains two major loops. The outer loop
is called an iteration, while the inner loop within the Differential Evolution (DE) is called a generation.
Since the metamodel evaluation is fast, many metamodel evaluations are possible during the DE process.
Usually, a very large number of generations (1000–10,000) can be performed on large populations
(40–100 individuals) within seconds.

Algorithm 1 Single-objective metamodel assisted differential evolution.

1: create an initial database

2: for it = 1 to required do
3: train metamodel on database
4: perform single-objective DE by using the metamodel
5: evaluate the best individual by expensive evaluation tool and add to the database
6: end for

The database stores the relationship between design vectors (input) and performance vectors (output)
of the designs already analyzed by the accurate evaluation. Several geometries are analyzed at the start of
the optimization algorithm by the accurate evaluation in order to train the first metamodel. The accuracy
of the Kriging predictions strongly depends on the information contained in the database. The Design of
Experiments (DOE) method maximizes the amount of information contained in the database for a limited
number of computations and is used in the present optimization algorithm [11]. The DOE technique
considers that each of the k design variables can take two values, fixed at 25% and 75% of the maximum
design range. The maximum amount of information can be established by evaluating each possible
combination of the design variables, which is called a full factorial design. This would yield a total of 2k
experiments, which becomes rapidly unfeasible for a larger number of design variables k. The current work
uses a 2k− p fractional factorial design, where only a fraction 1/2 p of the full factorial design is analyzed
with k − p = 7. One additional computation is made to evaluate the central design with all parameters
at 50% of their range. This results in a total of 27 + 1 = 129 evaluations.
Once the database is available and the first Kriging model has been trained, the algorithm searches
for the next sample to enrich the database. There are different strategies to select the next sample to
be analyzed. The simplest approach, labeled Prediction Minimization, selects the best design according
Int. J. Turbomach. Propuls. Power 2019, 4, 35 4 of 20

to the metamodel prediction. The newly evaluated individual which is added to the database improves
the accuracy of the model in the region where previously the evolutionary algorithm was predicting a
minimum. This feedback is the most essential part of the algorithm as it makes the system self-learning.

2.2. Kriging Metamodel

The mathematical form of a Kriging model has two parts, as shown in Equation (1). The first part is a
linear regression with an arbitrary number k of regression functions g j , which tries to catch the main trend
of the response.
k
fe(~x) = ∑ β j gj (~x) + Z(~x). (1)
j =1

For the ordinary Kriging, only one regression function is used (k = 1), namely the process mean
(β 1 g1 (~x) = µ̂). The second part of Equation (1), Z (~x), is a model of a Gaussian random process with zero
mean with a covariance of Z (~x) assumed to be [12]:

Cov( Z (~x1 ), Z (~x2 )) = σ2 R(~x1 ,~x2 ), (2)

with σ2 the process variance and R(~x1 ,~x2 ) a correlation function, which depends only on the distance ~x1 −~x2 .
The correlation function is chosen as a product of one-dimensional correlation functions:

d
R(~x1 ,~x2 ) = ∏ R(x1i , x2i ). (3)
i =1

The parameters β j and the function Z (~x) are determined such that fe(~x) is the best linear unbiased
predictor (BLUP). A linear estimator means that fe(~x) can be written as a linear combination of the
observation samples and the unbiasedness constraint means that the mean error of the approximation
is zero. The best linear unbiased predictor is considered the predictor with minimal mean square error
(MSE) of the predictions,  2 
MSE = E fe(~x) − f (~x) . (4)

The BLUP formulation when simplified for ordinary Kriging reads as follows [13]:

fe(~x) = µ̂ + r T (~x)R−1 (y − 1µ̂) (5)

with r(~x) the vector of the spatial correlation between the unknown point ~x and all known samples ~xi
h i
r T (~x) = R(~x,~x1 ) R(~x,~x2 ) ... R(~x,~xn ) , (6)

and finally R the correlation matrix which contains the evaluation of spatial correlation function at each
combination of samples:  
R(~x1 ,~x1 ) R(~x1 ,~x2 ) . . . R(~x1 ,~xn )
 R(~x2 ,~x1 ) R(~x2 ,~x2 ) . . . R(~x2 ,~xn ) 
 
R= .. .. .. .. . (7)
. . . .
 
 
R(~xn ,~x1 ) R(~xn ,~x2 ) ... R(~xn ,~xn )
Note that usually, the diagonal elements of the correlation matrix are equal to one, while off-diagonal
entries are between 0 and 1 and get closer to 1 when ~xi is closer to ~x j . The parameters used to build R
Int. J. Turbomach. Propuls. Power 2019, 4, 35 5 of 20

serve as tuning parameters to improve the prediction accuracy of the method. The correlation function can
take different forms: exponential, Cubic Spline, Matèrn, Gaussian, etc. [14] and the Gaussian correlation,
one of the most used correlations [6], reads as follows:

d
R(~x1 ,~x2 ) = exp(− ∑ θ (l ) || xi − x j ||2 ).
(l ) (l )
(8)
l =1

The hyperparameters θ l , the variance σ2 and the process mean µ̂ are tuning parameters of the Kriging
model and are found by maximizing the Likelihood Estimation (LE):

1
LE = p exp((y − Fµ̂) T R−1 (y − Fµ̂)). (9)
σ2 |R|(2π )n

It is possible to find the values of µ̂ and σ2 that maximize LE directly by setting the partial derivative
of the likelihood equal to zero:
F T R −1 y
µ̂ = ,
F T R −1 F (10)
1
σ̂2 = (y − Fµ̂) T R−1 (y − Fµ̂).
n
The value of θ that maximizes LE, however, cannot be calculated analytically and has to be obtained
numerically during the model training. Two training processes are established: Maximum Likelihood
Estimation (MLE) and Cross Validation (CR). MLE is faster but yields generally lower levels of accuracy [15]
as it assumes the samples behave as a Gaussian process. Cross validation provides better accuracy by
minimizing a leave-one-out error but is therefore very costly, especially for large training sets. In this work,
the training uses MLE on the concentrated log-likelihood φ instead of Equation (9), which simplifies the
LE expression but results in the same optimal θ:

n 1
φ = − ln(σ̂2 ) − ln(| R|). (11)
2 2
Typically, a bounded optimization problem is solved to find the θ hyperparameter from a predefined
interval [θlb –θub ] with the highest likelihood using the definition of the process variance and mean
of Equation (10). The lower bound θlb and the upper bound θub impact the quality of the kriging
interpolation. When θ → 0, the correlation matrix approaches a matrix of ones, as clear from Equation (8),
and the prediction is reduced to the mean value of all samples. On the other hand, for large values
of θ, the correlation matrix approaches the identity matrix and the prediction is still interpolating the
samples but equal to the mean value for any new prediction (see [16] for more details on the effect of the
hyperparameters).
For the bounded Kriging, the MLE-based training follows a two-step strategy. First, the hyperparameters
intervals are scanned to find a θ interval that guarantees an interpolation error lower than 1%. This step
guarantees a θlb value high enough to have an acceptable prediction error. When a nugget (a nugget or
regularization parameter is a small number added to the diagonal of the correlation matrix.) is used as in this
work, the check of the interpolation error guarantees also an acceptable θhb . Next, a search is done to find the
optimal d hyperparameters within the interval defined at Step 1. The MLE uses a gradient-based optimization
method (L-BFGS) from the NLOPT library, which performs a bounded hyperparameter optimization. The first
step of this procedure differentiate the bounded Kriging from ordinary Kriging, for the latter the lower and
upper bounds θlb and θub are set based on heuristic and trial-and-error.
Int. J. Turbomach. Propuls. Power 2019, 4, 35 6 of 20

2.3. Exploration Versus Exploitation

The metamodel-assisted optimization algorithm builds first an interpolation around the available
samples from the database. During subsequent iterations, the metamodel proposes new samples for the
high-fidelity evaluation. These samples refine gradually the models in certain areas of the objective space.
Two main trends are possible: either the model refines the area next to the optimum of the last iteration
or it explores areas where the model accuracy is still low. This step is called the infill criterion or acquisition.

2.3.1. Prediction Minimization (PM)

This criterion, especially suited for dense initial databases, chooses the design with the optimal
prediction (e.g., minimum loss or highest efficiency). A large initial database covering well the
multidimensional design space leads, in fact, to a good overall prediction accuracy. The learning is
in general very effective and the metamodel-assisted optimizer converges rapidly to an optimum, which is
likely to be a global one in case a global search algorithm is used.
In realistic settings, however, the initial database is not dense enough as the high-fidelity evaluation
is time-consuming and the required number of design evaluations scales up exponentially with the
number of design variables (curse of dimensionality). A pure exploitative infill criterion steers hence the
metamodel-assisted optimizer toward a local optimum when starting from sparse data and very likely
will not search in unexplored regions where a global minimum can be hidden (see Figure 1).

DB sample
Infill Sample at itr i
i
15
µ= +/- σ
10

-5 54 3 2 1

0.5 1
Figure 1. Minimization of the Forrester function with Minimal Prediction (MP) criteria. The algorithm is
trapped in a local optimum.

2.3.2. Maximum Entropy

Unlike the PM criterion, the Maximum Entropy criterion chooses the design with the maximum
uncertainty or variance ŝ2 (see Equation (12)) on the prediction and does not consult the objective value.

1 − F T R −1 r ( x )
s2 ( x ) = σ̂2 (1 − r ( x ) T R−1 r ( x ) + ). (12)
F T R −1 F
The algorithm is then space-filling and works on improving the general prediction accuracy without
using the system response of evaluated samples for the next infill. In practice, the metamodel-assisted
optimizer with a variance maximization criterion needs many high-fidelity evaluations, especially for
Int. J. Turbomach. Propuls. Power 2019, 4, 35 7 of 20

high-dimensional problems. Moreover, many explored areas are likely to be of low interest but need to be
investigated to ensure that no potential optimum is missed, which slows down the optimizer convergence.
It is more interesting to explore areas of the design space only when it is likely that they will improve the
objective function. The next criterion tackles this issue.

2.3.3. Expected Improvement (EI)

The EI criterion quantifies the expected improvement I in a certain location and the MAEA algorithm
uses the information to search for the location of the highest expected improvement.
To visualize the expected improvement, one needs to look at the probability density function for a
given x value, as shown in Figure 2. The mean value is given by µw while the the best value in the current
set of samples is shown by ymin . The probability that the prediction is equal to y for this x value is given by:

1 ( y − µ x )2
φ(y) = √ exp(− ). (13)
2πs x 2s( x )2

We are of course only interested in y-values that are smaller than ymin . We introduce a new coordinate
axis I that starts from ymin and goes further to smaller y-values. Each value I corresponds to a certain
improvement and has a probability of improvement equal to [14]:

1 (ymin − I − µ̂ x )2
φ( I ) = √ exp[− ] (14)
2πs 2s( x )2

The interest now goes to the mean value of I. Small values of I have a larger probability of occurring
but larger values of I can still occur with significant probability if the variance s x is large. The mean value
of the distribution φ( I ) is the expected improvement (EI) and is given by:
Z I =∞
E( I ( x )) = Iφ( I )dI. (15)
I =0
When integrated by part, the latter equation has following form:

E( I ( x )) = s[uΦ(u) + φ(u)] (16)

y −µ̂
where u = mins .
After rearrangement, the EI reads as follows:

ymin − ŷ( x ) y − ŷ( x )

E[ I ( x )] = (ymin − ŷ( x ))Φ( ) + ŝφ( min ) if ŝ > 0
ŝ( x ) ŝ( x ) (17)
E[ I ( x )] = 0 if ŝ = 0

Schonlau [17] proved that EI-based metamodel-assisted optimizers can find a global optimum after
a finite number of iterations. The EI value is zero at sampled locations and might increase further from
samples as the variance grows. These two factors result in a multimodal shape of the EI requiring special
care to find the location in the design space with the highest EI value. Figure 3 shows the convergence of
the EI criterion on the Forrester function in a few iterations, unlike the prediction maximization criterion.
Int. J. Turbomach. Propuls. Power 2019, 4, 35 8 of 20

prediction

10 ymin EI=∫ IΦ(I)dI

true function
Φ(I)
PI=∫Φ(I)dI
ymin
T
PI I
~ grey area 0 ∞
0.5 1

Figure 2. Visualization of the EI using the probability density function at position x.

15 15

10 10

Y 5 Y 5

0 0
1 2 1
-5 -5
3
0.5 1 0.5 1

EI EI

Figure 3. Minimization of the Forrester function using the Expected Improvement (EI) infill criteria.
The algorithm finds the global optimum.

2.3.4. Cyclic Infill Criteria

It is possible, in a sequential optimization framework, to switch between different infill criteria during
the iterative cycle. A cyclic search repeats the same cycle of acquisition, which is characterized by a number
of iterations and a predefined criterion for each of these iterations. One cycle may start with few explorative
samples, then gradually evolves towards fewer explorative samples and finally extremely exploitative
samples, similar to Sasena [15] (p. 84). A cycle, in this work, is composed of m EI iterations followed by n
PM iterations with (m + n) small enough to allow multiple cycles during the optimization under limited
computational budget. EI is responsible for the exploration of the design space to uncover promising
regions, while the prediction minimization (PM) exploits the best design the metamodel can offer.

2.4. Treating Failed Evaluations

In a metamodel assisted optimization, the high-fidelity evaluation could be a CFD simulation, a stress
analysis or any field-specific simulation. Simulations could diverge or even fail during the meshing phase,
especially when challenging designs are generated or the design parameters are not properly bound.
In that case, a decision has to be made on how to use the information from the failed simulation [10].
Ignoring the failed designs creates an information loss and the optimizer is likely to propose very
similar designs to the failing one. When noticing the failed simulation, it is possible to push the search out
of the region by an intrusive penalty strategy or a non-intrusive recovery strategy. Assigning a penalty for
the objective value of the failed design masks the failure region for the optimizer, so it does not approach
Int. J. Turbomach. Propuls. Power 2019, 4, 35 9 of 20

the same area again. The penalty distorts, however, the objective space and may hide an optimum if
located next to the area of the failed simulation.
The non-intrusive strategy, on the other hand, replaces the penalty with a dummy value generated
by the surrogate model [9]. Therefore, it is of utmost importance to use an explorative infill criterion,
otherwise the surrogate could get trapped in an endless loop of proposing the same design which fails.
The penalty and recovery approaches were tested on the Rastrigin analytical 2D function:
n
f ( x ) = 10 + ∑ [ xi2 − 10cos(2πxi )], xi ∈ [−5.12, 5.12] (18)
i =1

The Rasrigin function has a global shape influenced by the parabolic term x2 , which leads to the
global optimum at (0, 0). The sinusoidal term creates waves on top of the parabolic shape, which render
the optimization more difficult with many local optima. Moreover, the evaluation of the Rastrigin function
has been slightly changed and, for certain ranges of the variables x1 and x2 , the evaluation of the function
is set artificially to fail. Figure 4a shows a contour plot of the Rastrigin 2D function with a ring to simulate
failed evaluations surrounding the optimum in the center. Both presented approaches were tested and the
results are shown in Figure 4b. The penalty alternative distorts the objective space as expected and misses,
therefore, the global optimum. While the penalty approach is trapped in a local minimum, the recovery
approach is able to find the global optimum.

1
10
Penalty
Distance to Optimum

0
10

10-1

-2
10

10-3 Recovery
-4
10

100 200 300 400 500

ITR

(a) (b)

Figure 4. Penalty versus recovery: (a) Adapted Rastrigin case; and (b) Convergence history.

3. Application
The application is a highly constrained radial compressor for turbocharger applications. This section
introduces first the geometry of the test-cases and then defines the optimization problem specifying the
objectives and the constraints. Finally, the results of the meta-model assisted optimization are analyzed.

3.1. Geometry Definition and Design Variables

The 3D radial compressor is defined by:

• the meridional contour at hub and shroud (Figure 5);

• the camber line of the main blade (Figure 6a);
• the camber line of the splitter blade, as a deviation from the main blade;
• the thickness distribution for both main and splitter blade (Figure 6b); and
• the number of blades
Int. J. Turbomach. Propuls. Power 2019, 4, 35 10 of 20

Diffusor

TE

LE
Inlet
R

Figure 5. Geometry definition of the hub and shroud meridional flow path highlighting some design
variables related to the geometry.

The meridional shape of the impeller is defined by several Bézier curves or B-spline curves for both
hub and shroud contours. In the present application, the contours are subdivided into three main curves:
one for the inlet, one for the blade passage, and one for the diffuser. The coordinates of the control points
are geometrical parameters that can be changed by the optimization program and allow to modify the
meridional passage. In Figure 5, the control points that are considered as free parameters are depicted
schematically by arrows. For some of the control points, the coordinates are directly controlled by
the optimizer, while other coordinates are controlled by other parameters, e.g., angles and distances.
This parameterization has been introduced to easily respect tangency at each interface of two patches,
and to have better control on the different shapes that can be generated by the optimizer. The backplate
thickness is also controlled by one degree of freedom. In total, 11 parameters control the meridional shape;
for each parameter, upper and lower limits are defined.

Main Blade Hub

Main Blade Schroud

0 0.2 0.4 0.6 0.8 1

(a) (b)

Figure 6. Geometry definition of the camber line by a beta distribution and the blade thickness of a
radial compressor: (a) camber line definition; and (b) blade thickness definition.

The blade camber lines at hub and shroud are defined by the distribution of the angle β(u) between
the meridional plane m and the streamline S (Figure 6a). They are also defined by Bézier curves, where a
Int. J. Turbomach. Propuls. Power 2019, 4, 35 11 of 20

third-order curve is used for the hub and a fourth-order is used for the shroud (see Figure 7). The camber
line circumferential position θ (Figure 6a) is then defined by

R · dθ = dm · tan( β) (19)

and allows the transformation from the β distribution to ( x, y, z) coordinates of the camber line provided
that the meridional shape is given. In total, nine design parameters control the main blade camber line
definition at hub and shroud.

0 0.2 0.4 0.6 0.8 1

Blade Beta Distribution

Figure 7. Beta distribution for the main blade hub (red) and shroud (black).

When only a camber line is defined at the hub and shroud, the resulting blade will be a ruled surface
and the blade can be manufactured by flank-milling. If however an additional camber line is defined in
between the hub and shroud, an additional flexibility in geometry is tolerated, at the expense of a more
complicated manufacturing process. In the present work, one additional camber line at the mid-section
is defined. To avoid a parameterization which allows very distorted blade shapes, it is opted to define the
camber line as a difference relative to the camber line that results from a ruled surface approach. Four
additional design parameters control this additional flexibility.
For the splitter blade, the blade angle definition is defined relative to the main blade. A ∆β distribution,
defined by a second-order Bézier curve, is added to the main blade β-distribution to have good control
over the changes to the splitter blade. In addition, the position of the splitter blade is moved rearward
relative to the main blade, which is also free within this optimization. This represents two additional
design variables: one for the hub and one for the shroud. The total degrees of freedom for the splitter
blade is thus 8.
The relative position of the camber line at hub and shroud needs to be defined as well. Usually, the leading
edge of the hub camber line is positioned at θ = 0 in cylindrical coordinates. The shroud camber line can then
be freely chosen relative to the hub camber line. Two possibilities exist, depending on whether the leading
edge (lean) or trailing edge (rake) position is defined relative to the hub camber line. The present application
uses the rake as a parameter as it is defined at a higher radius than the lean, resulting in a smaller sensitivity.
Connecting the camber lines at hub and shroud results in an infinitely thin blade on which a thickness
distribution is added at hub and shroud to reach the final blade shape. The blade thickness distributions
at hub and shroud are defined by B-spline curves which define the evolution of the thickness along the
camber line from leading to trailing edge. The thickness distribution is specified for hub and shroud for
both main and splitter blade. The thickness at the hub is proportionally scaled by a scale factor, which
Int. J. Turbomach. Propuls. Power 2019, 4, 35 12 of 20

constitutes an additional degree of freedom. The number of blades could also be a design parameter
to be optimized but has been fixed for manufacturing reasons. This brings the total number of design
parameters to 34.
The fluid and solid domains are defined by the geometrical model. For the evaluation of the geometry
in terms of aerodynamics and structural mechanics, these domains need to be discretized by a finite volume
and finite element grid, respectively. The fluid domain is meshed with a structured grid, while, for the solid
domain, unstructured quadratic tetrahedral elements are used as a compromise between element quality
and automatic meshing. First, a grid in the meridional plane from hub to shroud is generated, with a
stretching near the walls. For each grid line in the meridional plane, a surface of revolution is constructed.
The intersection with the blade is computed, and finally a structured grid in this surface is generated by an
elliptic smoother. The collection of all grids on these meridional surfaces constructs the full 3D grid. First,
the bounding surfaces are meshed by using a Delaunay mesh procedure. The interior solid grid is then
constructed using a similar procedure in 3D. In Figure 8, a view on the solid and fluid grid is shown.

Figure 8. View on the mesh for both fluid and solid domains.

3.2. Geometry Evaluation

The aerothermal and stress evaluations are made in parallel. NUMECA Fine Turbo is used to calculate
the aerodynamic performance of the radial compressor. A structured grid with 3M cells is used for all
computations to guarantee a comparable accuracy for all the samples evaluated. All computations are
non-adiabatic with specified wall temperatures. For each geometry, three CFD computations are performed
in parallel for the three operating points shown in Figure 9. For Operating Points 1 and 2 near surge,
the mass flow is imposed at the outlet and for Operating Point 3, which is situated near choke, the base
pressure is specified.
The open-source code CalculiX is used for the stress calculation. Similar grids with 250,000 nodes and
160,000 elements are used for all samples.
The grid is refined in areas of stress concentrations, e.g., at the fillet connecting the blade to the hub.
A periodic boundary condition is applied, such that only a limited part of the geometry needs to be analyzed.
Int. J. Turbomach. Propuls. Power 2019, 4, 35 13 of 20

OP3
1.0 OP1

πtt
πtt (OP3)
0.5 OP2

.
m
. (OP3)
0 0.25 0.5 0.75 1.0 m
Figure 9. The considered three operating points shown on a performance map.

3.3. Optimization Problem

Present optimization is a multi-objective multidisciplinary constrained optimization (see Figure 9).
The objective consists in increasing the aerodynamic performance of the impeller, while different
aerodynamic and mechanical constraints are applied.

3.3.1. Objective
The objective is to maximize the total to total efficiency of the compressor at Operating Points 1 and 2.
The definition of the objective function is:

f (~x) = −(0.7ηttOP1 + 0.3ηttOP2 ), (20)

with
T2ti s − T1t
ηtt = . (21)
T2t − T1t

3.3.2. Constraints
Several constraints are imposed. The first constraint considers the maximum von Mises stresses due
to the centrifugal forces in the compressor and is activated if the stress exceeds a defined material constant.
Figure 10 shows the von Mises stress inside the impeller for a particular geometry. Other mechanical
constraints, related to the mass and inertia of the impeller, the position of the gravity center (not too far
from the supporting bearings) and the radial displacement of the blades due to the centrifugal forces, are
not important in the present application and have no corresponding constraint.

g1 (~x) = σmax − σallowable < 0 (22)

A second constraint concerns the choke mass flow. At low back pressure, the impeller needs to be able
to deliver a minimum mass flow imposed by the manufacturer. The chocking mass flow is determined
from the CFD computation at Operating Point 3.

g2 (~x) = ṁchoke − ṁdesired < 0 (23)

The third constraint requires that, for Operating Point 2, which uses a mass flow boundary condition,
the desired pressure ratio is reached. This is necessary because otherwise, the optimizer will unload the
blades to reach higher efficiency, and thus reduce the work of the compressor.

g3 (~x) = π Req − πOP2 < 0 (24)

Int. J. Turbomach. Propuls. Power 2019, 4, 35 14 of 20

A fourth constraint is imposed on the momentum of inertia I, which needs to be limited to allow for
good transient behavior of the rotor. The momentum of inertia is computed from the solid mesh.

g4 (~x) = I − IReq < 0 (25)

Figure 10. Von Mises stresses due to centrifugal forces in the impeller.

3.4. Results
In total, 129 designs were generated by the DOE process to build the initial database for the
metamodel-assisted optimization. For each design, a stress analysis and three CFD evaluations were
evaluated, with two CFD evaluations for the efficiency at Operating Points 1 and 2 and a third at
Operating Point 3 for the choking massflow. The DOE algorithm checked the results of the designs
evaluation for consistency and designs for which the evaluation was not successful (e.g., diverging CFD,
meshing problems) were automatically removed. In the current case, only a few designs were rejected.
It is worth to note that, in the initial database, not a single design satisfied all constraints.
The ordinary Kriging proposed a list of designs for which the CFD at Operating Point 3 was
failing. therefore all results below belong to the metamodel-assisted optimization using the proposed
bounded Kriging. For each optimization, two Kriging models were devoted to predicting the efficiency at
Operating Points 1 and 2, necessary for the objective, and four other Kriging models were used to predict
the constraints. The training used the Maximum Likelihood Estimation (MLE) and a gradient-based
method (L-BFGS) was used to maximize the likelihood. For this test case, Kriging used the cyclic infill
criterion explained in Section 2.3.4: for one iteration, the DE searched for the design maximizing the
prediction of the compressor efficiency and for the next the DE searched for the prediction maximizing the
Expected Improvement (EI) of the efficiency.
Figure 11 shows the Euclidean distance between a design and the next one of a subsequent iteration
for both infill criteria when using bounded Kriging. During the search for a feasible design which satisfies
all constraints, the optimizer moved slowly within the design space making small changes to the design
from one iteration to the next. Once a feasibility region was found around iteration 140, the optimizer
started exploring the design space, an exploration led by the EI criterion, which introduced more changes
to the design, reflected in the higher Euclidean distance between designs, as shown in Figure 11.
Int. J. Turbomach. Propuls. Power 2019, 4, 35 15 of 20

0 PM
10 EI

Distance to next iteration

-1
10

50 100 150 200 250

Design Iteration
Figure 11. Euclidean distance to next design of subsequent iteration for both the explorative infill (EI) and
the exploitative one (PM) when using bounded Kriging.

Figure 12 shows the evolution of the bounded kriging-based prediction of the objective versus the
CFD-based evaluation during the optimization. The Kriging model predicted the efficiency with an
average error of 0.08% for Operating Point 1 and 0.7% for Operating Point 2. The large difference between
both operating points can be explained by the higher variation in efficiency at Operating Point 2, which is
located closer to surge (see Figure 9).

Simulation
Kriging
3
Efficiency Improvement [%]

2.5

1.5
50 100 150 200
Iteration Number
Figure 12. Comparison of the CFD-based and bounded Kriging-based predictions for the efficiency
improvement at every iteration relative to the baseline design.
Int. J. Turbomach. Propuls. Power 2019, 4, 35 16 of 20

Figure 13 shows the evolution of the bounded Kriging-based prediction and the simulation-based
evaluation for the performance parameters used to evaluate the four constraints. The relatively high
prediction error of the stresses did not affect the metamodel-assisted optimizer because the stress-related
constraint was satisfied early in the optimization process, as shown in Figure 13. The mass flow at OP3
rose gradually to meet the required amount after 40 iterations while the pressure ratio at OP1 took 150
iterations to reach the required threshold value. The pressure ratio at OP1, the maximum moment of
inertia at OP3 and the choking massflow at OP3 were predicted with an average prediction error of 0.085%,
0.21% and 0.15%, respectively. The Von Mises stresses, however, had an average prediction error of 2.8%.

Baseline
Simulation Simulation Simulation
Kriging Kriging Kriging
Simulation
Simulation Baseline
Stresses

Kriging

at OP3
at OP3

Kriging

Stresses
Inertia

massflow
massflow
VonM ofises

Von Mises
Moment

Choking
Choking

50 100 150
Iteration Number
50 50 100 100 150 150 200 50 50 100 100 150 150 200
Iteration Number Iteration
Iteration Number
Number

Baseline
Choking massflow at OP3

OP3
1
atOP2
ratio at
massflow

Baseline Simulation
Kriging
Pressure

Simulation
Kriging
Simulation Simulation
Choking

Kriging Kriging

50 100 150
50 50 100 100 150 150 200 50 Iteration
100 Number 150 200
Iteration Number Iteration Number

Figure 13. Plot of the simulated against predicted constraint values for bounded Kriging: the constraint
gradually being satisfied and baseline design violating all constraints.

The prediction error of efficiency during the optimization shows no tendency of decline, which is
related to weak learning, as depicted in Figure 12. The weak learning is due to the highly constrained
problem as the optimizer explored the design space for feasibility area without intensively sampling a
certain area. The optimizer searched first for designs that satisfy the four constraints before starting to
improve the objective, which is confirmed by the gap appearing between prediction and CFD stating
starting at iteration 145. When the feasibility area was found, the Kriging model needed to be trained to
this new environment where little designs were initially present. The cyclic infill criteria also contributed
to the exploratory behavior of the optimizer, which is necessary to solve such a highly constrained
optimization problem.
Int. J. Turbomach. Propuls. Power 2019, 4, 35 17 of 20

As shown in Figure 14, an improvement of 2.59% in the efficiency was reached while the baseline
design did not comply with any constraints. This proves how efficient and robust the metamodel-assisted
design optimization is for highly constrained turbomachinery problems of more than 30 design variables
when using bounded Kriging.

2.60

Obj of Best accepted design

2.50

2.40

2.30

2.20

140 160 180 200

Design Iteration

Figure 14. Best design satisfying all constraints during the optimization with bounded Kriging.

The optimized design reached lower values of entropy generation and had smaller separation bubbles,
as depicted in Figure 15. The optimizer increased the thickness of both main and splitter blades to comply
with the stress-related constraint. The blade loading comparison at 10% and 90% span for the baseline and
the optimized designs at OP1 is shown in Figure 16. The loading near the tip increased quite substantially
due to an increased turning.

Baseline at OP1 Optimized at OP1

Larger separation
bubbles

Higher
entopy

Figure 15. Entropy contours and streamlines of baseline and optimized designs.

The meridional contour of both designs is shown in Figure 17 along with the relative total pressure at
the outlet section. The drop of the latter was proportional to the losses, which were mainly caused by the
flow separation and the optimized design reached higher values of total pressure at the outlet.
Int. J. Turbomach. Propuls. Power 2019, 4, 35 18 of 20

1.4 Main Blade 10% Span 1.4 Main Blade 10% Span
Splitter 10% Span Splitter 10% Span
Main Blade 90% Span Main Blade 90% Span
1.2 Splitter 10% Sapn 1.2 Splitter 10% Sapn
Isentropic Machnumber

Isentropic Machnumber
1 1

0.8 0.8

0.6 0.6

0.4 0.4

0.2 0.2

0 0
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
relative meridional distance relative meridional distance

(a) (b)

Figure 16. Isentropic Machnumber of both blade and splitter for baseline and optimized design at Operating
Point 1: (a) baseline; and (b) optimized.

Outlet
section
total pressure at OP1

125%
Optimized
(Optimized/Baseline
Normalized relative

Optimized
Baseline
Baseline
R

100%

Inlet

75%

x 50% 100%
Outlet section [%]
Figure 17. Comparison of baseline and optimized design: (left) the meridional contour; and (right) the
relative total pressure at outlet section.

4. Conclusions
This article shows the potential of metamodel-assisted algorithms for the optimization of highly
constrained realistic 3D applications when using bounded Kriging. While the ordinary Kriging stopped
prematurely because of a set of failing evaluations, bounded Kriging with cyclic infill was able to explore
the design space efficiently and identify the area of feasible designs. After a feasible area was found,
the MAEA algorithm succeeded in improving the global total-to-total efficiency by 2.59%.
Int. J. Turbomach. Propuls. Power 2019, 4, 35 19 of 20

Author Contributions: Conceptualization, M.H.A.; Funding acquisition, T.V.; Methodology, M.H.A.; Project
administration, T.V.; Supervision, T.V.; Writing—original draft, M.H.A.; and Writing—review and editing, M.H.A.
and T.V.
Funding: This research was partially funded by VLAIO grant number IWT-140068. The APC was funded by Euroturbo.
Acknowledgments: I acknowledge colleagues L. Mueller and C. Chahine for the fruitful discussions on the
results analysis. This work was partially funded from VLAIO in the framework of the SBO EUFORIA project (IWT-140068).
The support is gratefully acknowledged.
Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of the study;
in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish
the results.

Abbreviations
The following abbreviations are used in this manuscript:

CFD Computational Fluid Dynamics

OP Operating Point
MAEA Metamodel-Assisted Evolutionary Algorithm
DE Differential Evolution
EA Evolutionary Algorithm
DOE Design Of Experiment
L-BFGS Limited-memory Broyden–Fletcher–Goldfarb–Shanno (BFGS)
PM Prediction Minimization
EI Expected Improvement

References
1. Mueller, L.; Verstraete, T. Adjoint-Based Multi-Point and Multi-Objective Optimization of a Turbocharger
Radial Turbine. Int. J. Turbomach. Propuls. Power 2019, 4, 10.
2. Holland, J.H. Adaption in Natural and Artificial Systems; University of Michigan Press: Ann Arbor, MI, USA, 1975.
3. Matheron, G. Principles of Geostatistics. Econ. Geol. 1963, 58, 1246–1266.
4. Toal, D.J.; Keane, A.J. Efficient multipoint aerodynamic design optimization via cokriging. J. Aircr.
2011, 48, 1685–1695.
5. Brooks, C.J.; Forrester, A.; Keane, A.; Shahpar, S. Multi-fidelity design optimisation of a transonic compressor rotor.
In Proceedings of the 9th European Conf. Turbomachinery Fluid Dynamics and Thermodynamics, Istanbul, Turkey,
21–25 March 2011..
6. Martin, J.D.; Simpson, T.W. Use of Kriging Models to Approximate Deterministic Computer Models. AIAA J.
2005, 43, 853–863.
7. Bellary, S.A.I.; Samad, A.; Couckuyt, I.; Dhaene, T. A comparative study of kriging variants for the optimization
of a turbomachinery system. Eng. Comput. 2016, 32, 49–59.
8. Toal, D.J. A study into the potential of GPUs for the efficient construction and evaluation of Kriging models.
Eng. Comput. 2016, 32, 377–404.
9. Forrester, A.; Keane, A. Engineering Design via Surrogate Modelling: A Practical Guide; John Wiley & Sons: Hoboken,
NJ, USA, 2008.
10. Stork, J.; Friese, M.; Zaefferer, M.; Bartz-Beielstein, T.; Fischbach, A.; Breiderhoff, B.; Naujoks, B.; Tušar, T.
Open Issues in Surrogate-Assisted Optimization. In High-Performance Simulation-Based Optimization; Springer:
Berlin/Heidelberg, Germany, 2020; pp. 225–244.
11. Kostrewa, K.; Alsalihi, Z.; Van den Braembussche, R.A. Optimization of Radial Turbines by Means of Design
of Experiment. Tech. Rep. VKI-PR-2003-17, von Karman Institute for Fluid Dynamics: Sint-Genesius-Rode,
Belgium, 2003.
Int. J. Turbomach. Propuls. Power 2019, 4, 35 20 of 20

12. Williams, C.K.; Rasmussen, C.E. Gaussian Processes for Machine Learning; MIT Press: Cambridge, MA, USA, 2006;
Volume 2.
13. Sacks, J.; Welch, W.J.; Mitchell, T.J.; Wynn, H.P. Design and analysis of computer experiments. Stat. Sci. 1989, 4,
409–423.
14. Jones, D.R. A taxonomy of global optimization methods based on response surfaces. J. Glob. Optim. 2001, 21, 345–383.
15. Sasena, M.J. Flexibility and Efficiency Enhancements for Constrained Global Design Optimization with
Kriging Approximations. Ph.D. Thesis, University of Michigan, Ann Arbor, MI, USA, 2002.
16. Aissa, M.H.; Mueller, L.; Verstraete, T. Optimization of a turbine inlet guide vane by gradient-based and
metamodel-assisted methods. Int. J. Comput. Fluid Dyn. 2019, in press.
17. Schonlau, M. Computer Experiments and Global Optimization; University of Waterloo: Waterloo, ON, Canada, 1997.

c 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY-NC-ND) license (http://creativecommons.org/licenses/by-nc-nd/4.0/).