【优化求解】基于精英反向学习带扰动因子的混沌蚁狮算法(EOPCALO)求解单目标优化问题附matlab代码...

1 简介

针对蚁狮算法易陷入局部最优、收敛速度慢的缺点，本文提出了基于精英反向学习带扰动因子的混沌蚁狮算法。该算法首先通过对蚂蚁的随机游走公式引入扰动因子，有效提高了寻优精度，避免算法陷入局部最优，有效平衡了全局最优搜索与局部最优搜索; 对蚁狮的精英个体进行反向学习生成精英反向解，增加了种群多样性; 通过比较当前最优解与精英反向解，选出优异个体作为新的全局最优解; 最后在蚂蚁受轮盘赌选择蚁狮、精英蚁狮影响的随机游走公式中引入 Logistic 混沌映射，有效提高了算法的全局搜索能力和寻优精度。通过 7 个经典测试函数实验对比，就目前改进的蚁狮算法而言，本文提出的算法能有效提高寻优的精度与收敛速度、高效解决函数优化问题。

2 部分代码

%___________________________________________________________________%

%  Ant Lion Optimizer (ALO) source codes demo version 1.0           %

% You can simply define your cost in a seperate file and load its handle to fobj 

% The initial parameters that you need are:

%__________________________________________

% fobj = @YourCostFunction

% dim = number of your variables

% Max_iteration = maximum number of generations

% SearchAgents_no = number of search agents

% lb=[lb1,lb2,...,lbn] where lbn is the lower bound of variable n

% ub=[ub1,ub2,...,ubn] where ubn is the upper bound of variable n

% If all the variables have equal lower bound you can just

% define lb and ub as two single number numbers

% To run ALO: [Best_score,Best_pos,cg_curve]=ALO(SearchAgents_no,Max_iteration,lb,ub,dim,fobj)

function [min_value,Elite_antlion_fitness,Elite_antlion_position,Convergence_curve]=ALO(N,Max_iter,lb,ub,dim,fobj)

% Initialize the positions of antlions and ants

antlion_position=initialization(N,dim,ub,lb);

ant_position=initialization(N,dim,ub,lb);

% Initialize variables to save the position of elite, sorted antlions, 

% convergence curve, antlions fitness, and ants fitness

Sorted_antlions=zeros(N,dim);

Elite_antlion_position=zeros(1,dim);

Elite_antlion_fitness=inf;

Convergence_curve=zeros(1,Max_iter);

antlions_fitness=zeros(1,N);

ants_fitness=zeros(1,N);

% Calculate the fitness of initial antlions and sort them

for i=1:size(antlion_position,1)

    antlions_fitness(1,i)=fobj(antlion_position(i,:)); 

end

[sorted_antlion_fitness,sorted_indexes]=sort(antlions_fitness);

for newindex=1:N

     Sorted_antlions(newindex,:)=antlion_position(sorted_indexes(newindex),:);

end

Elite_antlion_position=Sorted_antlions(1,:);

Elite_antlion_fitness=sorted_antlion_fitness(1);

% Main loop start from the second iteration since the first iteration 

% was dedicated to calculating the fitness of antlions

Current_iter=2; 

while Current_iter<Max_iter+1

    % This for loop simulate random walks

    for i=1:size(ant_position,1)

        % Select ant lions based on their fitness (the better anlion the higher chance of catching ant)

        Rolette_index=RouletteWheelSelection(1./sorted_antlion_fitness);

        if Rolette_index==-1  

            Rolette_index=1;

        end

        % RA is the random walk around the selected antlion by rolette wheel

        RA=Random_walk_around_antlion(dim,Max_iter,lb,ub, Sorted_antlions(Rolette_index,:),Current_iter);

        % RA is the random walk around the elite (best antlion so far)

        [RE]=Random_walk_around_antlion(dim,Max_iter,lb,ub, Elite_antlion_position(1,:),Current_iter);

        ant_position(i,:)= (RA(Current_iter,:)+RE(Current_iter,:))/2; % Equation (2.13) in the paper          

    end

    for i=1:size(ant_position,1)  

        % Boundar checking (bring back the antlions of ants inside search

        % space if they go beyoud the boundaries

        Flag4ub=ant_position(i,:)>ub;

        Flag4lb=ant_position(i,:)<lb;

        ant_position(i,:)=(ant_position(i,:).*(~(Flag4ub+Flag4lb)))+ub.*Flag4ub+lb.*Flag4lb;  

        ants_fitness(1,i)=fobj(ant_position(i,:));        

    end

    % Update antlion positions and fitnesses based of the ants (if an ant 

    % becomes fitter than an antlion we assume it was cought by the antlion  

    % and the antlion update goes to its position to build the trap)

    double_population=[Sorted_antlions;ant_position];

    double_fitness=[sorted_antlion_fitness ants_fitness];

    [double_fitness_sorted I]=sort(double_fitness);

    double_sorted_population=double_population(I,:);

    antlions_fitness=double_fitness_sorted(1:N);

    Sorted_antlions=double_sorted_population(1:N,:);

    % Update the position of elite if any antlinons becomes fitter than it

    if antlions_fitness(1)<Elite_antlion_fitness 

        Elite_antlion_position=Sorted_antlions(1,:);

        Elite_antlion_fitness=antlions_fitness(1);

    end

    % Keep the elite in the population

    Sorted_antlions(1,:)=Elite_antlion_position;

    antlions_fitness(1)=Elite_antlion_fitness;

    % Update the convergence curve

    Convergence_curve(Current_iter)=Elite_antlion_fitness;

    % Display the iteration and best optimum obtained so far

    if mod(Current_iter,1)==0

        display(['At iteration ', num2str(Current_iter), ' the elite fitness is ', num2str(Elite_antlion_fitness)]);

    end

min_value(Current_iter)=Elite_antlion_fitness;

    Current_iter=Current_iter+1; 

end

3 仿真结果

4 参考文献

[1]王茜,何庆,林杰,杨荣莹.精英反向学习带扰动因子的混沌蚁狮算法[J].智能计算机与应用,2020,10(08):51-57.

博主简介：擅长智能优化算法、神经网络预测、信号处理、元胞自动机、图像处理、路径规划、无人机等多种领域的Matlab仿真，相关matlab代码问题可私信交流。

部分理论引用网络文献，若有侵权联系博主删除。