test_functions N-D Test Functions D

class Damavandi(dimensions=2)

Damavandi objective function.

This class defines the Damavandi global optimization problem. This is a multimodal minimization problem defined as follows:

f_{\text{Damavandi}}(x) = \left[ 1 - \lvert{\frac{
\sin[\pi (x_1 - 2)]\sin[\pi (x2 - 2)]}{\pi^2 (x_1 - 2)(x_2 - 2)}}
\rvert^5 \right] \left[2 + (x_1 - 7)^2 + 2(x_2 - 7)^2 \right]

Here, n represents the number of dimensions and x_i \in
[0, 14] for i = 1, ..., n.

Damavandi function

Two-dimensional Damavandi function


Global optimum: f(x) = 0.0 for x_i = 2 for i = 1, ..., n

Jamil, M. & Yang, X.-S. A Literature Survey of Benchmark Functions For Global Optimization Problems Int. Journal of Mathematical Modelling and Numerical Optimisation, 2013, 4, 150-194.


class Deb01(dimensions=2)

Deb 1 objective function.

This class defines the Deb 1 global optimization problem. This is a multimodal minimization problem defined as follows:

f_{\text{Deb01}}(x) = - \frac{1}{N} \sum_{i=1}^n \sin^6(5 \pi x_i)

Here, n represents the number of dimensions and x_i \in [-1, 1] for i = 1, ..., n.

Deb01 function

Two-dimensional Deb01 function


Global optimum: f(x_i) = 0.0. The number of global minima is 5^n that are evenly spaced in the function landscape, where n represents the dimension of the problem.

Jamil, M. & Yang, X.-S. A Literature Survey of Benchmark Functions For Global Optimization Problems Int. Journal of Mathematical Modelling and Numerical Optimisation, 2013, 4, 150-194.


class Deb03(dimensions=2)

Deb 3 objective function.

This class defines the Deb 3 global optimization problem. This is a multimodal minimization problem defined as follows:

f_{\text{Deb03}}(x) = - \frac{1}{N} \sum_{i=1}^n \sin^6 \left[ 5 \pi
\left ( x_i^{3/4} - 0.05 \right) \right ]

Here, n represents the number of dimensions and x_i \in
[0, 1] for i = 1, ..., n.

Deb03 function

Two-dimensional Deb03 function


Global optimum: f(x) = 0.0. The number of global minima is 5^n that are evenly spaced in the function landscape, where n represents the dimension of the problem.

Jamil, M. & Yang, X.-S. A Literature Survey of Benchmark Functions For Global Optimization Problems Int. Journal of Mathematical Modelling and Numerical Optimisation, 2013, 4, 150-194.


class Decanomial(dimensions=2)

Decanomial objective function.

This class defines the Decanomial function global optimization problem. This is a multimodal minimization problem defined as follows:

f_{\text{Decanomial}}(x) = 0.001 \left(\lvert{x_{2}^{4} + 12 x_{2}^{3}
+ 54 x_{2}^{2} + 108 x_{2} + 81.0}\rvert + \lvert{x_{1}^{10}
- 20 x_{1}^{9} + 180 x_{1}^{8} - 960 x_{1}^{7} + 3360 x_{1}^{6}
- 8064 x_{1}^{5} + 13340 x_{1}^{4} - 15360 x_{1}^{3} + 11520 x_{1}^{2}
- 5120 x_{1} + 2624.0}\rvert\right)^{2}

with x_i \in [-10, 10] for i = 1, 2.

Decanomial function

Two-dimensional Decanomial function


Global optimum: f(x) = 0 for x = [2, -3]

Gavana, A. Global Optimization Benchmarks and AMPGO


class Deceptive(dimensions=2)

Deceptive objective function.

This class defines the Deceptive global optimization problem. This is a multimodal minimization problem defined as follows:

f_{\text{Deceptive}}(x) = - \left [\frac{1}{n}
\sum_{i=1}^{n} g_i(x_i) \right ]^{\beta}

Where \beta is a fixed non-linearity factor; in this exercise, \beta = 2. The function g_i(x_i) is given by:

g_i(x_i) = \begin{cases}
- \frac{x}{\alpha_i} + \frac{4}{5} & \textrm{if} \hspace{5pt} 0 \leq x_i \leq \frac{4}{5} \alpha_i \\
\frac{5x}{\alpha_i} -4 & \textrm{if} \hspace{5pt} \frac{4}{5} \alpha_i \le x_i \leq \alpha_i \\
\frac{5(x - \alpha_i)}{\alpha_i-1} & \textrm{if} \hspace{5pt} \alpha_i \le x_i \leq \frac{1 + 4\alpha_i}{5} \\
\frac{x - 1}{1 - \alpha_i} & \textrm{if} \hspace{5pt} \frac{1 + 4\alpha_i}{5} \le x_i \leq 1 \\
\end{cases}

Here, n represents the number of dimensions and x_i \in
[0, 1] for i = 1, ..., n.

Deceptive function

Two-dimensional Deceptive function


Global optimum: f(x) = -1 for x_i = \alpha_i for i = 1, ..., n

Gavana, A. Global Optimization Benchmarks and AMPGO

Todo

this function was taken from the Gavana website. The following code is based on his code. His code and the website don’t match, the equations are wrong.


class DeckkersAarts(dimensions=2)

Deckkers-Aarts objective function.

This class defines the Deckkers-Aarts global optimization problem. This is a multimodal minimization problem defined as follows:

f_{\text{DeckkersAarts}}(x) = 10^5x_1^2 + x_2^2 - (x_1^2 + x_2^2)^2
+ 10^{-5}(x_1^2 + x_2^2)^4

with x_i \in [-20, 20] for i = 1, 2.

DeckkersAarts function

Two-dimensional DeckkersAarts function


Global optimum: f(x) = -24776.518242168 for x = [0, \pm 14.9451209]

Jamil, M. & Yang, X.-S. A Literature Survey of Benchmark Functions For Global Optimization Problems Int. Journal of Mathematical Modelling and Numerical Optimisation, 2013, 4, 150-194.

Todo

jamil solution and global minimum are slightly wrong.


class DeflectedCorrugatedSpring(dimensions=2)

DeflectedCorrugatedSpring objective function.

This class defines the Deflected Corrugated Spring function global optimization problem. This is a multimodal minimization problem defined as follows:

f_{\text{DeflectedCorrugatedSpring}}(x) = 0.1\sum_{i=1}^n \left[ (x_i -
\alpha)^2 - \cos \left( K \sqrt {\sum_{i=1}^n (x_i - \alpha)^2}
\right ) \right ]

Where, in this exercise, K = 5 and \alpha = 5.

Here, n represents the number of dimensions and x_i \in
[0, 2\alpha] for i = 1, ..., n.

DeflectedCorrugatedSpring function

Two-dimensional DeflectedCorrugatedSpring function


Global optimum: f(x) = -1 for x_i = \alpha for i = 1, ..., n

Gavana, A. Global Optimization Benchmarks and AMPGO

Todo

website has a different equation to the gavana codebase. The function below is different to the equation above. Also, the global minimum is wrong.


class DeJong5(dimensions=2)

De Jong Function No. 5

DeJong5 function

Two-dimensional DeJong5 function



class DeVilliersGlasser01(dimensions=4)

DeVilliers-Glasser 1 objective function.

This class defines the DeVilliers-Glasser 1 function global optimization problem. This is a multimodal minimization problem defined as follows:

f_{\text{DeVilliersGlasser01}}(x) = \sum_{i=1}^{24} \left[ x_1x_2^{t_i}
\sin(x_3t_i + x_4) - y_i \right ]^2

Where, in this exercise, t_i = 0.1(i - 1) and y_i = 60.137(1.371^{t_i}) \sin(3.112t_i + 1.761).

Here, n represents the number of dimensions and x_i \in
[1, 100] for i = 1, ..., 4.

Global optimum: f(x) = 0 for x_i = 0 for x = [60.137, 1.371, 3.112, 1.761].

Jamil, M. & Yang, X.-S. A Literature Survey of Benchmark Functions For Global Optimization Problems Int. Journal of Mathematical Modelling and Numerical Optimisation, 2013, 4, 150-194.


class DeVilliersGlasser02(dimensions=5)

DeVilliers-Glasser 2 objective function.

This class defines the DeVilliers-Glasser 2 function global optimization problem. This is a multimodal minimization problem defined as follows:

f_{\text{DeVilliersGlasser01}}(x) = \sum_{i=1}^{24} \left[ x_1x_2^{t_i}
\tanh \left [x_3t_i + \sin(x_4t_i) \right] \cos(t_ie^{x_5}) -
y_i \right ]^2

Where, in this exercise, t_i = 0.1(i - 1) and y_i = 53.81(1.27^{t_i}) \tanh (3.012t_i + \sin(2.13t_i))
\cos(e^{0.507}t_i).

with x_i \in [0.5, 60] for i = 1, ..., 5.

Global optimum: f(x) = 0 for x = [53.81, 1.27, 3.012, 2.13, 0.507].

Jamil, M. & Yang, X.-S. A Literature Survey of Benchmark Functions For Global Optimization Problems Int. Journal of Mathematical Modelling and Numerical Optimisation, 2013, 4, 150-194.


class DixonPrice(dimensions=2)

Dixon and Price objective function.

This class defines the Dixon and Price global optimization problem. This is a multimodal minimization problem defined as follows:

f_{\text{DixonPrice}}(x) = (x_i - 1)^2
+ \sum_{i=2}^n i(2x_i^2 - x_{i-1})^2

Here, n represents the number of dimensions and x_i \in [-10, 10] for i = 1, ..., n.

DixonPrice function

Two-dimensional DixonPrice function


Global optimum: f(x_i) = 0 for x_i = 2^{- \frac{(2^i - 2)}{2^i}} for i = 1, ..., n

Jamil, M. & Yang, X.-S. A Literature Survey of Benchmark Functions For Global Optimization Problems Int. Journal of Mathematical Modelling and Numerical Optimisation, 2013, 4, 150-194.

Todo

Gavana code not correct. i array should start from 2.


class Dolan(dimensions=5)

Dolan objective function.

This class defines the Dolan global optimization problem. This is a multimodal minimization problem defined as follows:

f_{\text{Dolan}}(x) = \lvert (x_1 + 1.7 x_2)\sin(x_1) - 1.5 x_3
- 0.1 x_4\cos(x_5 + x_5 - x_1) + 0.2 x_5^2 - x_2 - 1 \rvert

with x_i \in [-100, 100] for i = 1, ..., 5.

Global optimum: f(x_i) = 10^{-5} for x = [8.39045925, 4.81424707, 7.34574133, 68.88246895, 3.85470806]

Gavana, A. Global Optimization Benchmarks and AMPGO


class DropWave(dimensions=2)

DropWave objective function.

This class defines the DropWave global optimization problem. This is a multimodal minimization problem defined as follows:

f_{\text{DropWave}}(x) = - \frac{1 + \cos\left(12 \sqrt{\sum_{i=1}^{n}
x_i^{2}}\right)}{2 + 0.5 \sum_{i=1}^{n} x_i^{2}}

with x_i \in [-5.12, 5.12] for i = 1, 2.

DropWave function

Two-dimensional DropWave function


Global optimum: f(x) = -1 for x = [0, 0]

Gavana, A. Global Optimization Benchmarks and AMPGO

Previous topic

N-D Test Functions C

Next topic

N-D Test Functions E

This Page