和声搜索算法（带详细注释）MATLAB程序

% % Copyright (c) 2015, Yarpiz (www.yarpiz.com) % All rights reserved. Please read the "license.txt" for license terms. % % Project Code: YPEA117 % Project Title: Implementation of Harmony Search in MATLAB % Publisher: Yarpiz (www.yarpiz.com) % % Developer: S. Mostapha Kalami Heris (Member of Yarpiz Team) % % Contact Info: sm.kalami@gmail.com, info@yarpiz.com % clc; clear; close all; %% Problem Definition 问题定义 CostFunction=@(x) Sphere(x); % Cost Function 代价函数 nVar=9; % Number of Deciison Variables 决策变量数（这个是维度） VarSize=[1 nVar]; % Decision Variables Matrix Size 决策变量矩阵大小 VarMin=-10; % Decision Variables Lower Bound 决策变量下界 VarMax= 10; % Decision Variables Upper Bound 决策变量上界 %% Harmony Search Parameters 和声参数 MaxIt=5000; % Maximum Number of Iterations 最大迭代次数 HMS=100; % Harmony Memory Size 和声记忆库大小 nNew=1; % Number of New Harmonies 新和声数量（维度） HMCR=0.9; % Harmony Memory Consideration Rate 记忆取值概率 PAR=0.1; % Pitch Adjustment Rate 微调概率 FW=0.02*(VarMax-VarMin); % Fret Width (Bandwidth) BW FW_damp=0.995; % Fret Width Damp Ratio 带宽阻尼比 %% Initialization 初始化 % Empty Harmony Structure empty_harmony.Position=[]; empty_harmony.Cost=[]; % Initialize Harmony Memory HM=repmat(empty_harmony,HMS,1); % 举例：B = repmat(A, m, n) %将矩阵A复制m*n块，即B由m*n块A平铺而成（创建和声记忆库） % Create Initial Harmonies 创建初始和声 for i=1:HMS HM(i).Position=unifrnd(VarMin,VarMax,VarSize); %unifrnd产生一个VarSize大小的随机数矩阵，其元素在VarMin,VarMax之间均匀分布。 HM(i).Cost=CostFunction(HM(i).Position); %得到所有解的适应度 end % Sort Harmony Memory 排序（用适应度排序） [~, SortOrder]=sort([HM.Cost]); HM=HM(SortOrder); % Update Best Solution Ever Found 更新找到最佳方案（因为已排序，所以HM(1)就是最佳） BestSol=HM(1); % Array to Hold Best Cost Values 存储最佳成本值的数组 BestCost=zeros(MaxIt,1); %% Harmony Search Main Loop for it=1:MaxIt % Initialize Array for New Harmonies 建立一个空的NEW NEW=repmat(empty_harmony,nNew,1); % Create New Harmonies for k=1:nNew % Create New Harmony Position NEW(k).Position=unifrnd(VarMin,VarMax,VarSize); %赋值 for j=1:nVar if rand<=HMCR % Use Harmony Memory i=randi([1 HMS]); NEW(k).Position(j)=HM(i).Position(j); end % Pitch Adjustment if rand<=PAR %DELTA=FW*unifrnd(-1,+1); % Uniform经典 DELTA=FW*randn(); % Gaussian (Normal) 高斯 NEW(k).Position(j)=NEW(k).Position(j)+DELTA; end end % Apply Variable Limits NEW(k).Position=max(NEW(k).Position,VarMin); NEW(k).Position=min(NEW(k).Position,VarMax); % Evaluation NEW(k).Cost=CostFunction(NEW(k).Position); end % Merge Harmony Memory and New Harmonies HM=[HM NEW]; %#ok % Sort Harmony Memory 排序 [~, SortOrder]=sort([HM.Cost]); HM=HM(SortOrder); % Truncate Extra Harmonies 取HMS个和声 HM=HM(1:HMS); % Update Best Solution Ever Found 得到最优解 BestSol=HM(1); % Store Best Cost Ever Found储存有史以来最好结果 BestCost(it)=BestSol.Cost; % Show Iteration Information disp(['Iteration ' num2str(it) ': Best Cost = ' num2str(BestCost(it))]); % Damp Fret Width FW=FW*FW_damp; end %% Results figure; % plot(BestCost,'LineWidth',2); semilogy(BestCost,'LineWidth',2); xlabel('Iteration'); ylabel('Best Cost'); grid on;