the particle swarm optimization algorithm
TRANSCRIPT
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 1/18
The Particle Swarm
Optimization Algorithm
Nebojša Trpković [email protected]
10 th Dec 2010
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 2/18
Nebojša Trpković [email protected] Slide 2 of 18
Problem Definition
optimization of continuous nonlinear functions
↓
finding the best solution in problem space
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 3/18
Nebojša Trpković [email protected] Slide 3 of 18
Example
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 4/18
Nebojša Trpković [email protected] Slide 4 of 18
Importance
• function optimization
• artificial neural network training
• fuzzy system control
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 5/18
Nebojša Trpković [email protected] Slide 5 of 18
Existing Solutions
• Ant Colony (ACO) – discrete
• Genetic Algorithms (GA) – slow convergence
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 6/18
Nebojša Trpković [email protected] Slide 6 of 18
Particle Swarm Optimization
Very simple classification:
• a computational method• that optimizes a problem• by iteratively trying to improve a candidate solution
• with regard to a given measure of quality
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 7/18
Nebojša Trpković [email protected] Slide 7 of 18
Particle Swarm Optimization
Facts:
• developed by Russell C. Eberhart and James Kennedy in 1995
• inspired by social behavior of bird flocking or fish schooling
• similar to evolutionary techniques such as Genetic Algorithms (GA)
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 8/18
Nebojša Trpković [email protected] Slide 8 of 18
Particle Swarm Optimization
Benefits:
• faster convergence
• less parameters to tune
↓
• easier searching in very large problem spaces
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 9/18
Nebojša Trpković [email protected] Slide 9 of 18
Particle Swarm Optimization
Basic principle:
let particle swarm movetowards the best position in search space,remembering each particle’s best known position
and global (swarm’s) best known position
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 10/18
Nebojša Trpković [email protected] Slide 10 of 18
Velocity Change
xi – specific particlep i – particle’s (personal) best known position g – swarm’s (global) best known position vi – particle’s velocity
vi ← ω vi + φ prp(p i - xi) + φ grg(g - xi)inertia cognitive social
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 11/18
Nebojša Trpković [email protected] Slide 11 of 18
Position Change
xi – specific particlevi – particle’s velocity
xi ← xi + vi
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 12/18
Nebojša Trpković [email protected] Slide 12 of 18
AlgorithmFor each particle
Initialize particleEND
Do
For each particle
Calculate fitness valueIf the fitness value is better than the best personal fitness value in history, set current valueas a new best personal fitness value
End
Choose the particle with the best fitness value of all the particles, and if that fitness value isbetter then current global best, set as a global best fitness value
For each particleCalculate particle velocity according velocity change equationUpdate particle position according position change equation
End
While maximum iterations or minimum error criteria is not attained
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 13/18
Nebojša Trpković [email protected] Slide 13 of 18
Single Particle
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 14/18
Nebojša Trpković [email protected] Slide 14 of 18
Parameters selection
Different ways to choose parameters:
• proper balance between exploration and exploitation (avoiding premature convergence to a local optimum yet still ensuring a good rate ofconvergence to the optimum)
• putting all attention on exploitation(making possible searches in a vast problem spaces)
• automatization by meta-optimization
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 15/18
Nebojša Trpković [email protected] Slide 15 of 18
Avoiding Local Optimums
• adding randomization factor to velocity calculation
• adding random momentum in a specific iterations
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 16/18
Nebojša Trpković [email protected] Slide 16 of 18
Swarm
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 17/18
Nebojša Trpković [email protected] Slide 17 of 18
Conclusion
“This algorithm belongs ideologically to that philosophical school
that allows wisdom to emerge rather than trying to impose it,
that emulates nature rather than trying to control it,
and that seeks to make things simpler rather than more complex.”
James Kennedy, Russell Eberhart
8/13/2019 The Particle Swarm Optimization Algorithm
http://slidepdf.com/reader/full/the-particle-swarm-optimization-algorithm 18/18
Nebojša Trpković [email protected] Slide 18 of 18
References
• Wikipediahttp://www.wikipedia.org/
• Swarm Intelligencehttp://www.swarmintelligence.org/
• Application of a particle swarm optimization algorithm fordetermining optimum well location and type, Jerome Onwunaluand Louis J. Durlofsky, 2009
• Particle Swarm Optimization, James Kennedy and Russell Eberhart,1995http://www.engr.iupui.edu/~shi/Coference/psopap4.html
• Robot Swarm driven by Particle Swarm Optimizationalgorithm, thinkfluidhttp://www.youtube.com/watch?v=RLIA1EKfSys