Tags:
create new tag
, view all tags
In this technique, particles are initialized to a reference space with coordinates between [-1, 1]. The velocity of each particle is recalculated depending on the previous velocities of itself and the other particles, which in turn will determine its current position. An error function is used to compare distances between locations in order to choose the best position. The function is iterative and each particle maintains the position which has the lowest error value.

[EQ.4, EQ.5 with variable definitions]

Suggested parameters are POPSIZE = 10, c1 = 1.8, c2 = 1.8, wmax = 1.2, wmin = 0.1 and kmax = 100 as they provide faster convergence.

To ensure that calculations can be performed within the reference space [-1, 1], coordinates must be scaled down appropriately by a factor of [100 * hop count between p1 and p2] and subsequently, the final calculated position must be scaled up by the same amount.

-- XiaohongWu - 2012-01-24

Topic revision: r1 - 2012-01-24 - XiaohongWu
 
This site is powered by the TWiki collaboration platformCopyright © 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback