pso and its variants swarm intelligence group peking university
TRANSCRIPT
![Page 1: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/1.jpg)
PSO and its variants
Swarm Intelligence Group Peking University
![Page 2: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/2.jpg)
Outline Classical and standard PSO
PSO on Benchmark Function
Analysis of PSO_state of art
Analysis of PSO_our idea
variants of PSO_state of art
Our variants of PSO
Applications of PSO
![Page 3: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/3.jpg)
Classical and standard PSO
Swarm is better than personal
![Page 4: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/4.jpg)
Classical and standard PSO
Russ EberhartRuss Eberhart James KennedyJames Kennedy
![Page 5: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/5.jpg)
Classical
1 2() ( ) () ( ) (1)
(2)id id id id d id
id id id
V w V c Rand p x c Rand g x
x x V
Vid : Velocity of each particle in each dimension i: Particle D: Dimension
W : Inertia Weight c1、 c2 : Constants Rand() : Random Pid : Best position of each particle gd : Best position of swarm xid : Current position of each particle in each dimension
![Page 6: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/6.jpg)
Classical and standard PSO
1 2() ( ) () ( ) (1)
(2)id id id id d id
id id id
V w V c Rand p x c Rand g x
x x V
( )idx t
( )dg t
x
y
( )idp t( )idV t
( 1)idV t ( 1)idx t
![Page 7: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/7.jpg)
Flow chart depicting the General PSO Algorithm:
![Page 8: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/8.jpg)
simulation 1
x
y
fitnessmin
max
search space
![Page 9: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/9.jpg)
simulation 2
x
y
search space
fitnessmin
max
![Page 10: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/10.jpg)
simulation 3
x
y
fitnessmin
max
search space
![Page 11: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/11.jpg)
simulation 4
x
y
fitnessmin
max
search space
![Page 12: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/12.jpg)
simulation 5
x
y
fitnessmin
max
search space
![Page 13: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/13.jpg)
simulation 6
x
y
fitnessmin
max
search space
![Page 14: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/14.jpg)
simulation 7
x
y
fitnessmin
max
search space
![Page 15: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/15.jpg)
simulation 8
x
y
fitnessmin
max
search space
![Page 16: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/16.jpg)
Schwefel's function
n :1=i 420.9687,=
418.9829;=)(
minimum global
500500
where
)sin()()(1
i
i
n
iii
x
nxf
x
xxxf
![Page 17: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/17.jpg)
Evolution - Initialization
![Page 18: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/18.jpg)
Evolution - 5 iteration
![Page 19: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/19.jpg)
Evolution - 10 iteration
![Page 20: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/20.jpg)
Evolution - 15 iteration
![Page 21: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/21.jpg)
Evolution - 20 iteration
![Page 22: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/22.jpg)
Evolution - 25 iteration
![Page 23: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/23.jpg)
Evolution - 100 iteration
![Page 24: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/24.jpg)
Evolution - 500 iteration
![Page 25: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/25.jpg)
Search result
Iteration Swarm best0 416.245599
5 515.748796
10 759.404006
15 793.732019
20 834.813763
100 837.911535
5000 837.965771
Global 837.9658
![Page 26: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/26.jpg)
Standard benchmark functions
nn
ii xxxf 5,5,
1
2
1
1
2221 10,10,1100
n
i
niii xxxxxf
1) Sphere Function
2) Rosenbrock Function
D
iii xxxf
1
2 102cos10
3) Rastrigin Function
4) Ackley Function
nn
ii
n
i
xxn
xn
exf 32,32,2cos1
exp1
2.0exp202011
2
![Page 27: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/27.jpg)
Composition Function
![Page 28: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/28.jpg)
Analysis of PSO_state of art Stagnation - Convergence
Clerc 2002 The particle swarm - explosion, stability, and convergence in a multidimensio
nal complex space,2002 Kennedy 2005
Dynamic-Probabilistic Particle Swarms,2005 Poli 2007
Exact Analysis of the Sampling Distribution for the Canonical Particle Swarm Optimiser and its Convergence during Stagnation,2007
On the Moments of the Sampling Distribution of Particle Swarm Optimisers,2007
Markov Chain Models of Bare-Bones Particle Swarm Optimizers,2007
standard PSO Defining a Standard for Particle Swarm Optimization,2007
![Page 29: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/29.jpg)
Analysis of PSO_state of art standard PSO: constriction factor -
convergence Update formula
1 2() ( ) () ( )
id id id id d id
id id id
V w V c Rand p x c Rand g x
x x V
1 2( () ( ) () ( ) )
id id id id d id
id id id
V V c Rand p x c Rand g x
x x V
Equivalent
![Page 30: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/30.jpg)
Analysis of PSO_state of art standard PSO
50 particles Non-uniform initialization No evaluation when particle is out of
boundary
![Page 31: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/31.jpg)
Analysis of PSO_state of art standard PSO
A local ring topology
![Page 32: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/32.jpg)
Analysis of PSO_state of art How does PSO works?
Stagnation versus objective function Classical PSO versus Standard PSO Search strategy versus performance
![Page 33: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/33.jpg)
Classical PSO Main idea: Particle swarm optimization,1995
Exploit the current best position Pbest Gbest
Explore the unkown space
pbest gbest
![Page 34: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/34.jpg)
Classical PSO
Implementation
pbest
gbest
1 2() ( ) () ( ) (1)
(2)id id id id d id
id id id
V w V c Rand p x c Rand g x
x x V
pbest gbest
wV
![Page 35: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/35.jpg)
Analysis of PSO_our idea Search strategy of
PSO Exploitation Exploration
![Page 36: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/36.jpg)
Analysis of PSO_our idea Hybrid uniform distribution
pbest gbest
wV
wVExploitation
Exploration
![Page 37: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/37.jpg)
Analysis of PSO_our idea
( 1) ( ) ( )x t x t wV t Z
Sampling probability density-computable
![Page 38: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/38.jpg)
Analysis of PSO_our idea
![Page 39: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/39.jpg)
Analysis of PSO_our idea
![Page 40: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/40.jpg)
Analysis of PSO_our idea
wVSampling probability
![Page 41: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/41.jpg)
Analysis of PSO_our idea No inertia part(wV)
![Page 42: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/42.jpg)
Analysis of PSO_our idea Inertia part(wV)
![Page 43: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/43.jpg)
Analysis of PSO_our idea No inertia part(wV)
![Page 44: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/44.jpg)
Analysis of PSO_our idea Inertia part(wV)
![Page 45: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/45.jpg)
Analysis of PSO_our idea Difference among variants of PSO
Probability
Exploitation Exploration
Balance
![Page 46: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/46.jpg)
Analysis of PSO_our idea What is the property of the
iteration?
![Page 47: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/47.jpg)
Analysis of PSO_our idea Whether the search strategy is the same or whethe
r the PSO is adaptive when Same parameter(during the convergent process) Different parameter Different dimensions Different number of particles Different topology Different objective functions In different search phase(when slow or sharp slope,stagn
ation,etc) What’s the change pattern of the search strategy?
![Page 48: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/48.jpg)
Analysis of PSO_our idea What is the better PSO on the search strategy?
Simpler implement Using one parameter as a tuning knob instead of two in stan
dard PSO Prove they are equialent when setting some value of parame
ter Effective on most objective functions Adaptive
![Page 49: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/49.jpg)
Analysis of PSO_our idea Markov chain
State transition matrix
![Page 50: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/50.jpg)
Analysis of PSO_our idea Random process
Gaussian process Kernel mapping
Gauss process
Covarance matrix Kernel function
Mapping ability
Search straegy Effective?
Objective problem
![Page 51: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/51.jpg)
Analysis of PSO_our idea the object of our analysis
search strategy of PSO Different parameter sets In different dimensions Using different number of particles On different objective functions Fitness evaluation Different topology Markov or gauss process and kernel function
Direction to PSO Knob PSO
![Page 52: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/52.jpg)
Analysis of PSO_our idea
1
( )i if x ( )P Exploitation
65432
w:
c:
dim:
Num:
Fun:
Top:
1x
2x
3x
4x
5x
7xFEs: 6x 7
![Page 53: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/53.jpg)
Current results Variance with convergence
func_num=1; fes_num=5000; run_num=10; particles_num=50; dims_num=30;
![Page 54: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/54.jpg)
Current results Variance with dimensions
func_num=1; fes_num=3000; run_num=10; particles_num=50;
![Page 55: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/55.jpg)
Current results Variance with number of particles
func_num=1; fes_num=3000; run_num=10; dims_num=30;
![Page 56: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/56.jpg)
Current results
Variance with topology
![Page 57: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/57.jpg)
Current results
Variance with inertia weight
![Page 58: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/58.jpg)
Current results 1. Shifted Sphere Function 2. Shifted Schwefel's Problem 1.2
-100-50
050
100
-100
-50
0
50
100-1
0
1
2
3
4
5
x 104
-100-50
050
100
-100
-50
0
50
100-2
0
2
4
6
8
x 104
![Page 59: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/59.jpg)
PSO on Benchmark Function 3. Shifted Rotated High Conditioned Elliptic Function 4. Shifted Schwefel's Problem 1.2 with Noise in Fitness
-100-50
050
100
-100
-50
0
50
1000
1
2
3
4
x 1010
-100-80
-60-40
-200
0
50
1000
1
2
3
4
5
x 104
![Page 60: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/60.jpg)
Current results Variance with objective functions
Unimodal Functions Multimodal Functions Expanded Multimodal Functions Hybrid Composition Functions
![Page 61: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/61.jpg)
Current results Variance with objective functions
func_num=1,2,3,4; fes_num=3000; run_num=5; particles_num=50; dims_num=30;
![Page 62: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/62.jpg)
variants of PSO_state of art Traditional strategy
Simulated annealing Tabu strategy Gradient methods
Adopted from other fields Clonal operation Mutation operation
Heuristical Methods Advance and retreat
Structure topology Full connection Ring topology
![Page 63: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/63.jpg)
Our variants of PSO
CPSO AR-CPSO MPSO RBH-PSO FPSO
![Page 64: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/64.jpg)
Our variants of PSO
CPSO
n次迭代过后克隆保存的n个全局最优粒子
将所有克隆出来的粒子利用随机扰动变异
基于浓度机制的多样性保持策略进行选择操作
保存每一代的全局最优粒子作为第二步中克隆算子的父粒子
![Page 65: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/65.jpg)
Our variants of PSO
MPSO
![Page 66: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/66.jpg)
Our variants of PSO
AR-CPSO
![Page 67: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/67.jpg)
Our variants of PSO
FPSO
![Page 68: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/68.jpg)
Applications of PSO
![Page 69: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/69.jpg)
Applications of PSO
![Page 70: PSO and its variants Swarm Intelligence Group Peking University](https://reader035.vdocuments.us/reader035/viewer/2022062517/56649eda5503460f94be8ba2/html5/thumbnails/70.jpg)
Applications of PSO