一、获取代码方式
获取代码方式1：
通过订阅紫极神光博客付费专栏，凭支付凭证，私信博主，可获得此代码。
获取代码方式2：
完整代码已上传我的资源：【优化算法】群居蜘蛛优化算法（SSO）【含Matlab源码 1452期】
备注：
订阅紫极神光博客付费专栏，可免费获得1份代码（有效期为订阅日起，三天内有效）；
二、群居蜘蛛优化算法简介
群居蜘蛛算法 （Society Spider Algorithm， SSA） 是James 等人在 2015 年提出的一种智能启发式算法， 该算法通过模拟群居蜘蛛的觅食规律来实现寻优过程。 蜘蛛群由蜘蛛和蛛网组成， 蜘蛛网可以表示为优化问题的搜索空间， 蜘蛛所在位置表示可行的解决方案， 个体的好坏由适应度表征。 蜘蛛在网上行走会产生振动， 每个振动都会包含蜘蛛个体信息， 振动在蜘网中传播， 蜘蛛通过振动信息来判断自己与食物的位置信息， 进而实现最终的寻优目标。假设优化问题具有 n 维度搜索空间， 整个蜘蛛族群集合为 S， 种群数量为 N， 雌性和雄性蜘蛛数量分别表示为 Nf 和 Nm：
三、部分源代码
function [bfit,befit,spbest,spbesth] = SSO(spidn,itern)
%SCO Summary of this function goes here
% Detailed explanation goes here
%% Preliminares
%spidn, spiders number
%itern, iterations number
if nargin<1, spidn=50; end
if nargin<2, itern=500; end
%% Function
f=@Griewank;
xd=-600;
xu=600;
dims = 30;
for i=1:dims
lb(i,:)=xd;
ub(i,:)=xu;
end
%% Initial Parameters
rand('state',0'); % Reset the random generator
% Define the poblation of females and males
fpl = 0.65; % Lower Female Percent
fpu = 0.9; % Upper Female Percent
fp = fpl+(fpu-fpl)*rand; % Aleatory Percent
fn = round(spidn*fp); % Number of females
mn = spidn-fn; % Number of males
%Probabilities of attraction or repulsion
% Proper tunning for better results
pm = exp(-(0.1:(3-0.1)/(itern-1):3));
% Initialization of vectors
fsp = zeros(fn,dims); % Initlize females
msp = zeros(mn,dims); % Initlize males
fefit = zeros(fn,1); % Initlize fitness females
mafit = zeros(mn,1); % Initlize fitness males
spwei = zeros(spidn,1); % Initlize weigth spiders
fewei = zeros(fn,1); % Initlize weigth spiders
mawei = zeros(mn,1); % Initlize weigth spiders
%% Population Initialization
% Generate Females
for i=1:fn
fsp(i,1:dims)=lb(1)+rand(1,dims).*(ub(1)-lb(1));
end
% Generate Males
for i=1:mn
msp(i,1:dims)=lb(1)+rand(1,dims).*(ub(1)-lb(1));
end
%% **** Evaluations *****
% Evaluation of function for females
for i=1:fn
fefit(i)=f(fsp(i,:),dims);
end
% Evaluation of function for males
for i=1:mn
mafit(i)=f(msp(i,:),dims);
end
%% ***** Assign weigth or sort ***********
% Obtain weight for every spider
spfit = [fefit' mafit']'; % Mix Females and Males
bfitw = min(spfit); % best fitness
wfit = max(spfit); % worst fitness
for i=1:spidn
spwei(i) = 0.001+((spfit(i)-wfit)/(bfitw-wfit));
end
fewei = spwei(1:fn); % Separate the female mass
mawei = spwei(fn+1:spidn);% Separate the male mass
%% Memory of the best
% Check the best position
[~,Ibe] = max(spwei);
% Check if female or male
if Ibe > fn
% Is Male
spbest=msp(Ibe-fn,:); % Asign best position to spbest
bfit = mafit(Ibe-fn); % Get best fitness for memory
else
% Is Female
spbest=fsp(Ibe,:); % Asign best position to spbest
bfit = fefit(Ibe); % Get best fitness for memory
end
%% Start the iterations
for i=1:itern
%% ***** Movement of spiders *****
% Move Females
[fsp] = FeMove(spidn,fn,fsp,msp,spbest,Ibe,spwei,dims,lb,ub,pm(i));
% Move Males
[msp] = MaMove(fn,mn,fsp,msp,fewei,mawei,dims,lb,ub,pm(i));
%% **** Evaluations *****
% Evaluation of function for females
for j=1:fn
fefit(j)=f(fsp(j,:),dims);
end
% Evaluation of function for males
for j=1:mn
mafit(j)=f(msp(j,:),dims);
end
%% ***** Assign weigth or sort ***********
spfit = [fefit' mafit']'; % Mix Females and Males
bfitw = min(spfit); % best fitness
wfit = max(spfit); % worst fitness
% Obtain weight for every spider
for j=1:spidn
spwei(j) = 0.001+((spfit(j)-wfit)/(bfitw-wfit));
end
fewei = spwei(1:fn); % Separate the female mass
mawei = spwei(fn+1:spidn);% Separate the male mass
%% Mating Operator
[ofspr] = Mating(fewei,mawei,fsp,msp,dims);
%% Selection of the Mating
if isempty(ofspr)
% % Do nothing
else
[fsp,msp,fefit,mafit] = Survive(fsp,msp,ofspr,fefit,mafit,spfit,f,fn,dims);
% ***** Recalculate the weigth or sort ***********
spfit = [fefit' mafit']'; % Mix Females and Males
bfitw = min(spfit); % best fitness
wfit = max(spfit); % worst fitness
% Obtain weight for every spider
for j=1:spidn
spwei(j) = 0.001+((spfit(j)-wfit)/(bfitw-wfit));
end
fewei = spwei(1:fn); % Separate the female mass
mawei = spwei(fn+1:spidn);% Separate the male mass
end
%% Memory of the best
% Check if best position belongs to male or female
[~,Ibe2] = max(spwei);
if Ibe2 > fn
% Is Male
spbest2=msp(Ibe2-fn,:); % Asign best position to spbest
bfit2 = mafit(Ibe2-fn); % Get best fitness for memory
else
% Is Female
spbest2 = fsp(Ibe2,:); % Asign best position to spbest
bfit2 = fefit(Ibe2); % Get best fitness for memory
end
%% Global Memory
if bfit<=bfit2
bfit = bfit;
spbest = spbest; % Asign best position to spbest
befit(i) = bfit;
else
bfit = bfit2;
spbest = spbest2; % Asign best position to spbest
befit(i) = bfit;
end
spbesth(i,:)=spbest;
% %% Plot Results
plot(fsp(:,1),fsp(:,2),'r.',msp(:,1),msp(:,2),'bx',spbest(:,1),spbest(:,2),'go');
hold on
axis([lb(1) ub(1) lb(2) ub(2)])
drawnow
hold off
end
%% Display of results
figure
plot(befit)
end四、运行结果
五、matlab版本及参考文献
1 matlab版本
2014a
2 参考文献
[1] 包子阳,余继周,杨杉.智能优化算法及其MATLAB实例（第2版）[M].电子工业出版社，2016.
[2]张岩,吴水根.MATLAB优化算法源代码[M].清华大学出版社，2017.
[3]施媛波.基于改进的群居蜘蛛优化云计算任务调度算法[J].电脑编程技巧与维护. 2021,(04)
