Seminar was given at University of Zakho as a computational Intelligent (Lecture) by Younis Mohammed Younis (Younis Duhoky) from Duhok/Kurdistan region/Iraq *-*-*-*-*-*-*-*-*-*-*-*-* as a Reference, All Element in this Seminar are belonged to (Krishnanand N. Kaipa, Debasish Ghose auth. Glowworm Swarm Optimization Theory, Algorithms, and Applications)

1. 1
By

2. 1- What is GSO And Algorithm Main Idea
2-History
3-Algorithm Phases
4-Flowchart
5-Luciferin Update Phase
6- Movement Phase
7-Neighborhood range update Phase
8-Psoudo Code Of GSO Algorithm
9-Applications

3. Glowworm Swarm Optimization (GSO)
algorithm
The main idea :
3

4. Introduced by Krishnanand N.
Kaipa and Debasish Ghose
in 2005
History
4

5. “Luciferin Update
Phase”
luciferin quantity proportional
to the fitness of its current
location in the objective
function space.
“Movement
Phase”
During this phase each
glowworm decides using a
probabilistic mechanism to move
toward a neighbor that has
luciferin value higher than its
won.
“Neighborhood Range
Update Phase”
GSO uses and adaptive
neighborhood range in order to
detect the presence of multiple
peaks .
Algorithm
Phases
Each cycle of Algorithm Consists Three Phases
5

6. Flowchart
GSOGSO starts by placing a population of n
glowworms randomly in the search space so that
they are well dispersed. Initially, all the glowworms
contain an equal quantity of luciferin 0.
The luciferin update depends on the function
value at the glowworm position.
Glowworms has only two possible directions of
movement.
a chosen neighborhood range would work relatively better
on objective functions where the minimum inter-peak
distance.

7. 7 7
Phase1
The luciferin update rule is
Luciferin Update Phase
Represent the luciferin level associated with
glowworm i at time t . ℓ0 is (5)
p is the luciferin decay constant (0< p <1)
(0.4).
ƴ is the luciferin enhancement constant (0< ƴ <1)
(0.6).
J (𝒙𝒊(𝒕 + 𝟏)) Represent the value of the objective
function at glowworm i ‘s location at time t .

8. 8
The Probability of Movement for i to j
Phase2
Movement Phase A
Ni(t) Probability of Movement for i to j
j⋲ Ni(t)
Ni(t)={ j : di j (t) < r
𝑖
𝑑
(t) ; ℓi (t) < ℓj (t)} is the set of
neighbors of glowworm i at time t .
di j (t) represents the distance between
glowworms i and j at time t .
r
𝑖
𝑑
(t) represents the variable neighborhood range
associated with glowworm i at time t .

9. 9
The Movement from i location toward j location
Phase2
Movement Phase B
xi (t) ∈ Rm is the location of glowworm i , at time
t, in the m-dimensional real space Rm.
∥ . ∥ represents the Euclidean norm operator.
s (s>0) is the step size. (0.08)

10. 10
β is a constant parameter (0.8)
nt is a parameter used to control the number of
neighbors. (5)
Phase3
Neighborhood range update
Phase

11. Pseudo Code
For
GSO
Algorithm 11

12. Application
1-(Knapsack problem)
2-(Dispatching system of public transport)
3-(Multimodal Function with collective robotics)
4-(Chasing Multiple Mobile signal Sources)
5-(Rolling bearing fault diagnosis method)

13. 13
GSO