?

?1 基于tent混沌改進(jìn)粒子群

為改善基本粒子群優(yōu)化算法的尋優(yōu)性能,通過(guò)算法混合,在粒子群優(yōu)化算法中遙步引入優(yōu)進(jìn)策略和混沌搜索機(jī)制,以加強(qiáng)粒子群的局部尋優(yōu)效率和全局尋優(yōu)性能.井將粒子分為兩類,分別執(zhí)行不同的進(jìn)化機(jī)制,實(shí)現(xiàn)協(xié)同尋優(yōu),從而構(gòu)建為一種新的混沌混合粒子群優(yōu)化算法.標(biāo)準(zhǔn)測(cè)試函數(shù)的仿真優(yōu)化結(jié)果表明,該混合算法對(duì)較大規(guī)模的復(fù)雜問(wèn)題具有較強(qiáng)的求解能力.算法尋優(yōu)效率高,全局性能好,優(yōu)化結(jié)果穩(wěn)定,性能明顯優(yōu)于標(biāo)準(zhǔn)粒子群優(yōu)化算法以及遺傳算法等單一的隨機(jī)搜索方法.

?2 部分代碼

```matlab

%_________________________________________________________________________%

% 基于Tent混沌映射改進(jìn)的粒子群優(yōu)化算法? ? ? ? ? ? ?%

%_________________________________________________________________________%

% 使用方法

%__________________________________________

% fobj = @YourCostFunction? ? ? ? 設(shè)定適應(yīng)度函數(shù)

% dim = number of your variables? ?設(shè)定維度

% Max_iteration = maximum number of generations 設(shè)定最大迭代次數(shù)

% SearchAgents_no = number of search agents? ?種群數(shù)量

% lb=[lb1,lb2,...,lbn] where lbn is the lower bound of variable n? 變量下邊界

% ub=[ub1,ub2,...,ubn] where ubn is the upper bound of variable n? ?變量上邊界

clear all

clc

close all

SearchAgents_no=10; % 種群數(shù)量

Function_name='F4'; % Name of the test function that can be from F1 to F23 (Table 1,2,3 in the paper) 設(shè)定適應(yīng)度函數(shù)

Vmax=5;%速度上限

Vmin=-5;%速度下限

Max_iteration=50;% 進(jìn)化次數(shù)

a=0.5;%混沌系數(shù)

c1 = 1.49445;%個(gè)體學(xué)習(xí)率

c2 = 1.49445;%群體學(xué)習(xí)率

% Load details of the selected benchmark function

[lb,ub,dim,fobj]=Get_Functions_details(Function_name);? %設(shè)定邊界以及優(yōu)化函數(shù)

%lb%粒子最小值

%ub%粒子最大值

%dim%粒子維數(shù)

[Best_pos_tent,pso_curve_tent,Best_score_tent]=psoNew(SearchAgents_no,Max_iteration,lb,ub,dim,fobj,Vmax,Vmin,a,c1,c2); %tent混沌粒子群

[Best_pos,pso_curve,Best_score]=pso(SearchAgents_no,Max_iteration,lb,ub,dim,fobj,Vmax,Vmin,c1,c2); %基本粒子群

figure('Position',[269? ?240? ?660? ?290])

%Draw search space

subplot(1,2,1);

func_plot(Function_name);

title('Parameter space')

xlabel('x_1');

ylabel('x_2');

zlabel([Function_name,'( x_1 , x_2 )'])

%Draw objective space

subplot(1,2,2);

plot(pso_curve_tent,'Color','r')

hold on;

plot(pso_curve,'Color','b')

title('Objective space')

xlabel('Iteration');

ylabel('Best score obtained so far');

axis tight

grid on

box on

legend('Tent混沌策略pso','基本粒子群')

%

display(['The best solution obtained by Tent混沌策略PSO is : ', num2str(Best_pos_tent)]);

display(['The best optimal value of the objective funciton found by Tent混沌策略PSO is : ', num2str(Best_score_tent)]);

display(['The best solution obtained by PSO is : ', num2str(Best_pos)]);

display(['The best optimal value of the objective funciton found by Tent混沌策略PSO is : ', num2str(Best_score)]);

```


?3 仿真結(jié)果

4 參考文獻(xiàn)

[1]程志剛, 張立慶, 李小林,等. 基于Tent映射的混沌混合粒子群優(yōu)化算法[J]. 系統(tǒng)工程與電子技術(shù), 2007(01):103-106.