George Evers, MSE

Electrical Engineer with Programming Emphasis

MATLAB PSO Research Toolbox

Product Summary

· Specify the desired PSO algorithm.

· Watch particles fly through the search space.

· Enjoy quick execution.

· Minimize hard coding by setting all parameters in the control panel.

· Automatically validate inputs.

· Automatically increment benchmarks with each new set of trials.

· Automatically generate statistics for each set of trials conducted.

· Automatically organize statistics over multiple sets of trials into tables.

· Automatically title figures.

· Automatically generate legends.

· Auto-save figures in any format.

· Specify which data to retain in order to reduce memory consumption and maximize execution speed.

· Incrementally save data for reconstruction upon completion in order to further improve speed and avoid memory overflow!

· Automatically save workspaces using descriptive, transparent file names including time stamps.

· Automatically estimate completion time with excellent accuracy!

· Free yourself from the computer screen by listening for an automatic “choo choo” sound upon completion.

· The code is thoroughly commented.

The control panel allows direct control of PSO parameters and toolbox settings.  Select from algorithms such as Gbest PSO, Lbest PSO, RegPSO, GCPSO, MPSO, and OPSO via simple switches.  Activate or de-activate time-varying inertia weight; asymmetric initialization; velocity clamping; velocity reset to prevent violation of boundary constraints; an array of graph types; histories of important variables; and much more.  Specify the number of trials, maximum number of function evaluations or iterations per trial, contour map quality, font size, line width, and much more — with values recommended for your convenience.

 

I designed the toolbox to be highly scalable and suitable for testing new ideas.  This is the PSO toolbox I use for publishing!

Particles (numbering optional) can be watched traversing the contour map (color optional) of the benchmark selected.  Personal and global bests can be monitored via optional black tags.

To assist with thesis study, I developed an extensive MATLAB PSO toolbox suitable for industrial, academic, and personal use.  The product documentation includes a quick walk through to help you get started.

Download Toolbox & Documentation (v20100825i)
Subscribe to Update Notifications
Toolbox Update History | Documentation Update History

To contact me:
E-mail: george at georgeevers.org

(@ removed to avoid email harvesters)