Particle swarm optimization: Difference between revisions

Line 4:

</p>

<p>

The goal of parameter selection is to ensure that the global minimum is discriminated from any local minima, and that the minimum is accurately determined, and that convergence is achieved with acceptible resource usage. To provide a common basis for comparing implementations, the following test cases ~~and parameter sets~~ are recommended:

The goal of parameter selection is to ensure that the global minimum is discriminated from any local minima, and that the minimum is accurately determined, and that convergence is achieved with acceptible resource usage. To provide a common basis for comparing implementations, the following test cases are recommended:

<ul>

<li> McCormick function - bowl-shaped, with a single minimum;

<li> McCormick function - bowl-shaped, with a single minimum

<ul> function parameters and bounds (recommended):

recommended parameters:

omega = 0, phi p = 0.6, phi g = 0.3, number of particles = 100, number of iterations = 40 </li>

<li> Michalewicz function - steep ridges and valleys, with multiple minima; recommended parameters: omega = phi p = phi g = 0.3, number of particles = 1000, number of iterations = 30 </li>

</ul>

<ul> search parameters (suggested):

<li> omega = 0 </li>

<li> number of particles = 100 </li>

<li> number of iterations = 40 </li>

</ul>

<li> Michalewicz function - steep ridges and valleys, with multiple minima

<ul> function parameters and bounds (recommended):

</ul>

<ul> search parameters (suggested):

<li> omega = 0.3 </li>

<li> number of particles = 1000 </li>

<li> number of iterations = 30 </li>

</ul>

</p>