基于混合生物地理学的优化萤火虫算法研究（Matlab代码实现）

 🍒🍒🍒欢迎关注🌈🌈🌈

📝个人主页：我爱Matlab

👍点赞➕评论➕收藏 == 养成习惯（一键三连）🌻🌻🌻

🍌希望大家多多支持🍓~一起加油 🤗

💬语录：将来的我一定会感谢现在奋斗的自己！

🍁🥬🕒摘要🕒🥬🍁

针对传统多目标算法在应用中解决多目标优化问题时存在的Pareto前沿收敛不好、解集均匀性差等问题,文章通过实例对混合生物地理学算法（HBBO）进行研究和综合分析,将其与优化萤火虫算法相互结合来求解多目标优化的问题。

✨🔎⚡运行结果⚡🔎✨

​

 

 

 

💂♨️👨‍🎓Matlab代码👨‍🎓♨️💂

%% Making Things Ready
clc;
clear;
warning('off');

%% Music Signal Loading
[signal,fs] = audioread('Setar.wav');
win = 0.050;
step = 0.050;
fs=44100;

%% Time Domain Features
EnergyEntropy = Energy_Entropy_Block(signal, win*fs, step*fs, 10)';
ShortTimeEnergy = ShortTimeEnergy(signal, win*fs, step*fs);

%% Frequency Domain Features
SpectralCentroid = SpectralCentroid(signal, win*fs, step*fs, fs);
SpectralFlux = SpectralFlux(signal,win*fs, step*fs, fs);

%% Making Inputs and Targets
Inputs=[EnergyEntropy SpectralCentroid SpectralFlux]';
Targets=ShortTimeEnergy';
data.Inputs=Inputs;
data.Targets=Targets;
data=JustLoad(data);

%% Generate Basic Fuzzy Model
ClusNum=3; % FCM Cluster Number
fis=GenerateFuzzy(data,ClusNum);

%% BBO-FireFly Algorithm Learning
BBOFuzzy=BBOFCN(fis,data);  
BBOFireFlyFuzzy=FireFlyFCN(BBOFuzzy,data);   

%% BBO-FireFly Results 
% BBO-FireFly 
TrTar=data.TrainTargets;
TrainOutputs=evalfis(data.TrainInputs,BBOFireFlyFuzzy);
% Basic and BBO-FireFly Models
BasicFeature=data.TrainTargets;
BBOFireFly=TrainOutputs;
% BBO-FireFly Train Errors Calculations
Errors=data.TrainTargets-TrainOutputs;
MSE=mean(Errors.^2);
RMSE=sqrt(MSE);  
error_mean=mean(Errors);
error_std=std(Errors);

%% BBO-FireFly Algorithm Plots
% Plot Input Signal
figure('units','normalized','outerposition',[0 0 1 1])
subplot(4,1,1);
plot(signal);
title('Input Audio Signal');
grid on;
% Plot Train Result
subplot(4,1,2);
plot(data.TrainTargets,'--',...
'LineWidth',2,...
'MarkerSize',3,...
'MarkerEdgeColor','b',...
'Color',[0.0,0.0,0.9]);
hold on;
plot(TrainOutputs,'-',...
'LineWidth',2,...
'MarkerSize',3,...
'MarkerEdgeColor','m',...
'Color',[0.9,0.9,0.0]);
legend('Basic Model','BBO-FireFly Model');
title('BBO-FireFly Signal Trained');
xlabel('Sample Index');
grid on;
% Plot Distribution Fit Histogram
subplot(4,1,3);
h=histfit(Errors, 80);
h(1).FaceColor = [.3 .7 0.7];
title([' BBO-FireFly Train Error  =   ' num2str(RMSE)]);
% Plot Signals
subplot(4,1,4);
plot(normalize(EnergyEntropy),'-^');hold on;
plot(normalize(SpectralCentroid),'-o');hold on;
plot(normalize(SpectralFlux),'-d');hold on;
plot(normalize(ShortTimeEnergy),'-s');hold on;
plot(normalize(BBOFireFly),'-*');
hold off;
legend('Energy Entropy','Spectral Centroid', 'Spectral Flux', 'Short Time Energy', 'BBO-FireFly');
title('All Signals');
grid on;

%% Regression Line
[population2,gof] = fit(BasicFeature,BBOFireFly,'poly4');
figure;
plot(BasicFeature,BBOFireFly,'o',...
    'LineWidth',1,...
    'MarkerSize',8,...
    'MarkerEdgeColor','g',...
    'MarkerFaceColor',[0.9,0.2,0.2]);
    title(['BBO FireFly Train - R =  ' num2str(1-gof.rmse)]);
        xlabel('Train Target');
    ylabel('Train Output');   
hold on
plot(population2,'b-','predobs');
    xlabel('Train Target');
    ylabel('Train Output'); 
    grid on;
hold off;
 

📜📢🌈参考文献🌈📢📜

[1]张丹丽,高彦杰.多目标的混合生物地理学优化算法研究[J].科技创新与应用,2022,12(01):21-23+27.DOI:10.19981/j.CN23-1581/G3.2022.01.005.