A Unified Frame of Swarm Intelligence Optimization Algorithm
Swarm intelligence optimization algorithm is a heuristic search algorithm based on the swarm intelligent behavior of biology, which shows excellent performance in deal with complex optimization problems. On the basis of analyzing particle swarm optimizati
- PDF / 165,695 Bytes
- 7 Pages / 429.725 x 659.895 pts Page_size
- 8 Downloads / 209 Views
stract. Swarm intelligence optimization algorithm is a heuristic search algorithm based on the swarm intelligent behavior of biology, which shows excellent performance in deal with complex optimization problems. On the basis of analyzing particle swarm optimization algorithm, ant colony algorithm and artificial bee colony algorithm, the paper presents a unified frame of swarm intelligence optimization algorithm that is helpful for improving and perfecting swarm intelligence optimization algorithm. Keywords: swarm intelligence optimization algorithm, particle swarm optimization algorithm, ant colony algorithm, artificial bee colony algorithm, unified frame.
1 Introduction Swarm intelligence optimization algorithm stemmed from simulation of the evolution and swarm food-seeking living colony in nature. By studying living colony, people found that social living, such as ant, bird, fish and etc., shows unparalleled excellence in swarm than in single in food seeking or nest building, i.e., they shows great intelligence through some mechanism in swarm. Drawing inspiration from that, people design many optimization algorithm simulating colony living, such as ant colony algorithm[3], particle swarm optimization algorithm[1], shuffled frog-leaping algorithm[4], artificial bee colony algorithm[2] and etc., which are successfully applied in many fields. Different kinds of swarm intelligence optimization algorithms have different steps and processes, but their substances are quite similar. To study their common regularities and unified expression is helpful in improving and optimizing the algorithms, and helps to design new algorithms.
2 Swarm Intelligence Optimization Algorithm 2.1 Particle Swarm Optimization Algorithm Particle swarm optimization (PSO in short) algorithm, stemmed from simulation of the food-seeking behavior of bird flock, was presented by Eberhart and Kennedy in 1995 [1]. H. Tan (Ed.): Knowledge Discovery and Data Mining, AISC 135, pp. 745–751. © Springer-Verlag Berlin Heidelberg 2012 springerlink.com
746
J. Chen, Y. Zhang, and Y. Luo
In PSO algorithm, the solutions of a problem are supposed to a swarm of particles, and every particle has a fitness value determined by a function to be optimized and a speed to determine its flying direction and distance. The particles follow the current optimal particle to search in the solution space. PSO initializes a flock of random particles (i.e. random solution), and searches the optimal solution by iteration updating. In every iteration, a particle updates itself by tracking two extrema: one is the current optimal position found by the particle itself which is called individual extremum (pbest for short), and the other is the optimal position found by the whole flock which is called global extremum (gbest for short). At last, the optimal solution of the problem is found after many time’s iteration. The velocity and position are updated by the following equations at iteration t : v id (t + 1) = wv id (t ) + c1 r1 [ p id − x id (t )] + c 2 r2 [ p gd − x id (t )]
(1)
Data Loading...
