1 Content Introduction
This article briefly analyzes the current research status of swarm intelligence optimization algorithms, with a detailed description of the “Teaching-Learning” optimization algorithm, and analyzes the performance, advantages, and disadvantages of the “Teaching-Learning” algorithm. Several improved “Teaching-Learning” optimization algorithms are then introduced, and the application research situation of the “Teaching-Learning” optimization algorithm is discussed. Finally, the current problems existing in the “Teaching-Learning” optimization algorithm are explained, and future research directions for the “Teaching-Learning” optimization algorithm are pointed out.
2 Simulation Code
%
%------------------------------------------clc;clear;close all;warning('off');%%-----------------------------------------model=CreateM(); % Create Modelmodel.Umax=100;CostFunction=@(xhat) MyCost(xhat,model); % Cost FunctionVarSize=[model.K model.H]; % Size of Decision Variables MatrixnVar=prod(VarSize); % Number of Decision VariablesVarMin=0; % Lower Bound of VariablesVarMax=1; % Upper Bound of Variables%% TLBO ParametersMaxIt = 250; % Maximum Number of IterationsnPop = 150; % Population Size%% Start% Empty Structure for Individualsempty_individual.Position = [];empty_individual.Cost = [];empty_individual.Sol=[];% Initialize Population Arraypop = repmat(empty_individual, nPop, 1);% Initialize Best SolutionBestSol.Cost = inf;% Initialize Population Membersfor i = 1:nPoppop(i).Position = unifrnd(VarMin, VarMax, VarSize);[pop(i).Cost, pop(i).Sol]= CostFunction(pop(i).Position);if pop(i).Cost < BestSol.CostBestSol = pop(i);endend% Initialize Best Cost RecordBestCosts = zeros(MaxIt, 1);%% TLBO Bodyfor it = 1:MaxIt% Calculate Population MeanMean = 0;for i = 1:nPopMean = Mean + pop(i).Position;endMean = Mean/nPop;% Select TeacherTeacher = pop(1);for i = 2:nPopif pop(i).Cost < Teacher.CostTeacher = pop(i);endend% Teacher Phasefor i = 1:nPop% Create Empty Solutionnewsol = empty_individual;% Teaching FactorTF = randi([1 2]);% Teaching (moving towards teacher)newsol.Position = pop(i).Position ...+ rand(VarSize).*(Teacher.Position - TF*Mean);% Clippingnewsol.Position = max(newsol.Position, VarMin);newsol.Position = min(newsol.Position, VarMax);% Evaluation[newsol.Cost, newsol.Sol]= CostFunction(newsol.Position);% Comparisionif newsol.Cost<pop(i).Costpop(i) = newsol;if pop(i).Cost < BestSol.CostBestSol = pop(i);endendend% Learner Phasefor i = 1:nPopA = 1:nPop;A(i) = [];j = A(randi(nPop-1));Step = pop(i).Position - pop(j).Position;if pop(j).Cost < pop(i).CostStep = -Step;end% Create Empty Solutionnewsol = empty_individual;% Teaching (moving towards teacher)newsol.Position = pop(i).Position + rand(VarSize).*Step;% Clippingnewsol.Position = max(newsol.Position, VarMin);newsol.Position = min(newsol.Position, VarMax);% Evaluation[newsol.Cost, newsol.Sol]= CostFunction(newsol.Position);% Comparisionif newsol.Cost<pop(i).Costpop(i) = newsol;if pop(i).Cost < BestSol.CostBestSol = pop(i);endendend% Store Record for Current IterationBestCosts(it) = BestSol.Cost;% Show Iteration Informationdisp(['In Iteration ' num2str(it) ': TLBO Best Cost Is = ' num2str(BestCosts(it))]);figure(1);PlotSol(BestSol.Sol,model);pause(0.01);endtitle('TLBO Optimal Inventory Control');%% Plotfigure;semilogy(BestCosts,'k', 'LineWidth', 2);xlabel('Iteration');ylabel('Best Cost');grid on;3 Running Results
4 References
[1] Cheng Y. A Study on the Scheduling Problem of Fuzzy Flexible Job Shop Based on Teaching-Learning Optimization Algorithm. Journal of Xinxiang University, 2021, 38(9):6.
[2] Yang W., Gu X. Research on Rescheduling and Inventory Optimization of Continuous Production Process Based on Chaotic Optimization Algorithm. Journal of East China University of Science and Technology: Natural Science Edition, 2006, 32(7):4.