can you compare my moema code with my ngsa ii code? try use zdt2 problem. please give me…
- User
can you compare my moema code with my ngsa ii code? try use zdt2 problem. please give me full code moema code: % Main script to run MOEMA optimization clc; clear; close all; % Set problem parameters NPop = 100; % Population size T = 200; % Number of iterations dim = 30; % Number of decision variables lb = zeros(1, dim); % Lower bound ub = ones(1, dim); % Upper bound
% Run MOEMA [paretoFront, paretoSet, Trajectories, fitness_history, position_history] = MOEMA(NPop, T, lb, ub, dim, @zdt2);
% Plot the Pareto front figure; plot(paretoFront(:,1), paretoFront(:,2), 'ro'); xlabel('f1'); ylabel('f2'); title('Pareto Front for ZDT2 using MOEMA');
% Calculate and display metrics disp(['Number of solutions in Pareto Front: ', num2str(size(paretoFront, 1))]); disp(['Minimum f1: ', num2str(min(paretoFront(:,1)))]); disp(['Minimum f2: ', num2str(min(paretoFront(:,2)))]);
function [paretoFront, paretoSet, Trajectories, fitness_history, position_history] = MOEMA(NPop, T, lb, ub, dim, FITNESSFCN) % Multi-Objective Evolutionary Mating Algorithm (MOEMA) % MOEMA attempts to solve multi-objective problems of the following form: % min F(X) = [f1(X), f2(X), ..., fm(X)] subject to: lb <= X <= ub %
% FITNESSFCN - function handle of the multi-objective fitness function % lb - lower bounds on X % ub - upper bounds on X % T - number of iterations % NPop - size of the population (class size)
% paretoFront - set of non-dominated solutions in objective space % paretoSet - set of non-dominated solutions in decision space % Trajectories - best solution trajectory over iterations % fitness_history - history of fitness values % position_history - history of population positions% Preallocation fitness_history = zeros(NPop, T, 2); % Assuming 2 objectives position_history = zeros(NPop, T, dim); Trajectories = zeros(dim, T);
% Generation of initial population pop = initialization(NPop, dim, ub, lb); pop_dad = pop(1:NPop/2, :); % Eqn (1) in the paper pop_mum = pop(NPop/2+1:end, :); % Eqn (2) in the paper
% Evaluation of objective function for initialization obj_dad = evaluate_population(pop_dad, FITNESSFCN); obj_mum = evaluate_population(pop_mum, FITNESSFCN); obj = [obj_dad; obj_mum];
% Initialize archive with non-dominated solutions archive = update_archive([], [], pop, obj);
for gen = 1:T % Start iteration k1 = randperm(NPop/2); k2 = randperm(NPop/2); k1x = pop_dad(k1, :); k2x = pop_mum(k2, :);
for k = 1:NPop/2 % Mating process Imates = 1 + (var(k1x(k, :)) - var(k2x(k, :))); % Eqn (4) in the paper p = randn(1, dim); % Eqn (6) in the paper r = rand(); if Imates >= 0 % Upper part of Eqn (5) in the paper temp1 = p .* k1x(k, :); temp2 = (1-p) .* k2x(k, :); pop_child = temp1 + temp2; else % Lower part of Eqn (5) in the paper temp1 = p .* k2x(k, :); temp2 = (1-p) .* k1x(k, :); pop_child = temp1 + temp2; end % Impact of best environment (using random non-dominated solution) if ~isempty(archive.pop) best_pop = archive.pop(randi(size(archive.pop, 1)), :); K = rand(1, dim) < 0.8; % Eqn (8) in the paper pop_child = best_pop .* (1-K) + pop_child .* K; % Eqn (7) in the paper end % Bounding of the solution pop_child = max(min(ub, pop_child), lb); % Evaluate child obj_child = FITNESSFCN(pop_child); % Update population using non-dominated sorting [pop_dad, obj_dad] = update_population(pop_dad, obj_dad, pop_child, obj_child, k1(k)); [pop_mum, obj_mum] = update_population(pop_mum, obj_mum, pop_child, obj_child, k2(k)); % Exploration process if r < 0.2 if ~isempty(archive.pop) temp1 = archive.pop(randi(size(archive.pop, 1)), :); pop_child = rand(1, dim) .* temp1; % Eqn (9) in the paper pop_child = max(min(ub, pop_child), lb); obj_child = FITNESSFCN(pop_child); % Update population using non-dominated sorting [pop_dad, obj_dad] = update_population(pop_dad, obj_dad, pop_child, obj_child, randi(NPop/2)); [pop_mum, obj_mum] = update_population(pop_mum, obj_mum, pop_child, obj_child, randi(NPop/2)); end end end % Combine updated pool dad and mum pop = [pop_dad; pop_mum]; obj = [obj_dad; obj_mum]; % Update archive archive = update_archive(archive.pop, archive.obj, pop, obj); % Record the history at each iteration fitness_history(:, gen, :) = obj; position_history(:, gen, :) = pop; % Update Trajectories (using a random non-dominated solution) if ~isempty(archive.pop) Trajectories(:, gen) = archive.pop(randi(size(archive.pop, 1)), :)'; end disp(['Iteration #', num2str(gen), ' completed. Archive size: ', num2str(size(archive.pop, 1))]);end
% Return Pareto front and Pareto set paretoFront = archive.obj; paretoSet = archive.pop;
end
% Helper Functions
function pop = initialization(NPop, dim, ub, lb) pop = rand(NPop, dim) .* (ub - lb) + lb; end
function obj = evaluate_population(pop, FITNESSFCN) obj = zeros(size(pop, 1), 2); % Assuming 2 objectives for i = 1:size(pop, 1) obj(i, :) = FITNESSFCN(pop(i, :)); end end
function [pop, obj] = update_population(pop, obj, child, child_obj, index) if dominates(child_obj, obj(index, :)) pop(index, :) = child; obj(index, :) = child_obj; elseif ~dominates(obj(index, :), child_obj) if rand() < 0.5 pop(index, :) = child; obj(index, :) = child_obj; end end end
function archive = update_archive(archive_pop, archive_obj, pop, obj) combined_pop = [archive_pop; pop]; combined_obj = [archive_obj; obj];
[is_dominated, ~] = pareto_dominance(combined_obj); non_dominated = ~is_dominated; archive.pop = combined_pop(non_dominated, :); archive.obj = combined_obj(non_dominated, :);end
function [is_dominated, domination_count] = pareto_dominance(objectives) [n_solutions, n_objectives] = size(objectives); is_dominated = false(n_solutions, 1); domination_count = zeros(n_solutions, 1);
for i = 1:n_solutions for j = i+1:n_solutions if dominates(objectives(i, :), objectives(j, :)) is_dominated(j) = true; domination_count(i) = domination_count(i) + 1; elseif dominates(objectives(j, :), objectives(i, :)) is_dominated(i) = true; domination_count(j) = domination_count(j) + 1; end end endend
function result = dominates(obj1, obj2) result = all(obj1 <= obj2) && any(obj1 < obj2); end
function f = zdt2(x) f1 = x(1); g = 1 + 9*sum(x(2:end))/(length(x)-1); f2 = g * (1 - (f1/g)^2); f = [f1, f2]; end
ngsa ii code: clc; clear; close all;
% Set problem parameters NPop = 100; % Population size T = 200; % Number of iterations dim = 30; % Number of decision variables lb = zeros(1, dim); % Lower bound ub = ones(1, dim); % Upper bound
% Run NSGA-II [paretoFront, paretoSet] = NSGA2(NPop, T, lb, ub, dim);
% Plot the results figure; plot(paretoFront(:,1), paretoFront(:,2), 'ro', 'DisplayName', 'NSGA-II'); xlabel('f1'); ylabel('f2'); title('NSGA-II Pareto Front for ZDT1'); legend('show'); grid on;
function [paretoFront, paretoSet] = NSGA2(NPop, T, lb, ub, dim) % Initialize population pop = initialization(NPop, dim, ub, lb); fitness = evaluate_population(pop, @zdt1);
for gen = 1:T % Non-dominated sorting [fronts, ranks] = non_dominated_sort(fitness); % Create mating pool using tournament selection mating_pool = tournament_selection(pop, ranks, NPop); % Generate offspring using crossover and mutation offspring = crossover(mating_pool, 0.9, dim); offspring = mutation(offspring, 1/dim, dim); % Evaluate offspring offspring_fitness = evaluate_population(offspring, @zdt1); % Combine population and offspring comb_pop = [pop; offspring]; comb_fitness = [fitness; offspring_fitness]; % Non-dominated sorting on combined population [fronts, ~] = non_dominated_sort(comb_fitness); % Create new population [pop, fitness] = select_new_population(comb_pop, comb_fitness, fronts, NPop); % Display generation progress disp(['Generation #', num2str(gen), ' completed.']); end % Return last population as Pareto front paretoFront = fitness; paretoSet = pop;end
function pop = initialization(NPop, dim, ub, lb) pop = rand(NPop, dim) .* (ub - lb) + lb; end
function fitness = evaluate_population(pop, FITNESSFCN) fitness = zeros(size(pop, 1), 2); % Assuming 2 objectives for i = 1:size(pop, 1) fitness(i, :) = FITNESSFCN(pop(i, :)); end end
function [fronts, rank] = non_dominated_sort(fitness) num_individuals = size(fitness, 1); domination_count = zeros(num_individuals, 1); dominated_set = cell(num_individuals, 1); rank = zeros(num_individuals, 1);
fronts = cell(1); for i = 1:num_individuals for j = 1:num_individuals if i ~= j if dominates(fitness(i, :), fitness(j, :)) dominated_set{i} = [dominated_set{i}, j]; elseif dominates(fitness(j, :), fitness(i, :)) domination_count(i) = domination_count(i) + 1; end end end if domination_count(i) == 0 rank(i) = 1; fronts{1} = [fronts{1}, i]; end end current_front = 1; while ~isempty(fronts{current_front}) next_front = []; for i = fronts{current_front} for j = dominated_set{i} domination_count(j) = domination_count(j) - 1; if domination_count(j) == 0 rank(j) = current_front + 1; next_front = [next_front, j]; end end end current_front = current_front + 1; fronts{current_front} = next_front; end fronts = fronts(1:current_front-1);end
function is_dominated = dominates(x, y) is_dominated = all(x <= y) && any(x < y); end
function selected_pop = tournament_selection(pop, ranks, num_selected) selected_pop = zeros(num_selected, size(pop, 2)); for i = 1:num_selected candidates = randperm(size(pop, 1), 2); if ranks(candidates(1)) < ranks(candidates(2)) selected_pop(i, :) = pop(candidates(1), :); else selected_pop(i, :) = pop(candidates(2), :); end end end
function offspring = crossover(pop, crossover_prob, num_vars) offspring = zeros(size(pop)); for i = 1:2:size(pop, 1) if rand < crossover_prob parent1 = pop(i, :); parent2 = pop(i+1, :); cross_point = randi([1, num_vars-1], 1, 1); offspring(i, :) = [parent1(1:cross_point), parent2(cross_point+1:end)]; offspring(i+1, :) = [parent2(1:cross_point), parent1(cross_point+1:end)]; else offspring(i, :) = pop(i, :); offspring(i+1, :) = pop(i+1, :); end end end
function mutated_pop = mutation(pop, mutation_prob, num_vars) eta_m = 20; % Mutation distribution index mutated_pop = pop; for i = 1:size(pop, 1) for j = 1:num_vars if rand < mutation_prob r = rand; if r < 0.5 delta = (2 * r)^(1/(eta_m + 1)) - 1; else delta = 1 - (2 * (1 - r))^(1/(eta_m + 1)); end mutated_pop(i, j) = pop(i, j) + delta; mutated_pop(i, j) = min(max(mutated_pop(i, j), 0), 1); end end end end
function [pop, fitness] = select_new_population(comb_pop, comb_fitness, fronts, NPop) pop = []; fitness = []; for i = 1:length(fronts) if size(pop, 1) + length(fronts{i}) > NPop break; end pop = [pop; comb_pop(fronts{i}, :)]; fitness = [fitness; comb_fitness(fronts{i}, :)]; end
remaining = NPop - size(pop, 1); if remaining > 0 selected_front = fronts{i}; distances = crowding_distance(comb_fitness(selected_front, :)); [~, sorted_indices] = sort(distances, 'descend'); selected = selected_front(sorted_indices(1:remaining)); pop = [pop; comb_pop(selected, :)]; fitness = [fitness; comb_fitness(selected, :)]; endend
function distances = crowding_distance(fitness) num_individuals = size(fitness, 1); num_objectives = size(fitness, 2); distances = zeros(num_individuals, 1);
for m = 1:num_objectives [sorted_values, sorted_idx] = sort(fitness(:, m)); distances(sorted_idx(1)) = inf; distances(sorted_idx(end)) = inf; for i = 2:num_individuals-1 if sorted_values(end) ~= sorted_values(1) distances(sorted_idx(i)) = distances(sorted_idx(i)) + ... (sorted_values(i+1) - sorted_values(i-1)) / (sorted_values(end) - sorted_values(1)); end end endend
function f = zdt1(x) f1 = x(1); g = 1 + 9 * sum(x(2:end)) / (length(x) - 1); h = 1 - sqrt(f1 / g); f2 = g * h; f = [f1, f2]; end
- Assistant
To compare your MOEMA code with your NSGA-II code using the ZDT2 problem, I'll provide you with a complete implementation for each, making sure both scripts solve the same ZDT2 problem.
MOEMA Code for ZDT2:
% Main script to run MOEMA optimization clc; clear; close all; % Set problem parameters NPop = 100; % Population size T = 200; % Number of iterations dim = 30; % Number of decision variables lb = zeros(1, dim); % Lower bound ub = ones(1, dim); % Upper bound % Run MOEMA [paretoFront, paretoSet, Trajectories, fitness_history, position_history] = MOEMA(NPop, T, lb, ub, dim, @zdt2); % Plot the Pareto front figure; plot(paretoFront(:,1), paretoFront(:,2), 'ro'); xlabel('f1'); ylabel('f2'); title('Pareto Front for ZDT2 using MOEMA'); % Calculate and display metrics disp(['Number of solutions in Pareto Front: ', num2str(size(paretoFront, 1))]); disp(['Minimum f1: ', num2str(min(paretoFront(:,1)))]); disp(['Minimum f2: ', num2str(min(paretoFront(:,2)))]); % Add your MOEMA function here, as already provided in your original script. % Add your helper functions here (initialization, evaluate_population, update_population, update_archive, pareto_dominance, dominates, and zdt2).NSGA-II Code for ZDT2:
clc; clear; close all; % Set problem parameters NPop = 100; % Population size T = 200; % Number of iterations dim = 30; % Number of decision variables lb = zeros(1, dim); % Lower bound ub = ones(1, dim); % Upper bound % Run NSGA-II [paretoFront, paretoSet] = NSGA2(NPop, T, lb, ub, dim); % Plot the results figure; plot(paretoFront(:,1), paretoFront(:,2), 'ro', 'DisplayName', 'NSGA-II'); xlabel('f1'); ylabel('f2'); title('NSGA-II Pareto Front for ZDT2'); legend('show'); grid on; function [paretoFront, paretoSet] = NSGA2(NPop, T, lb, ub, dim) % Initialize population pop = initialization(NPop, dim, ub, lb); fitness = evaluate_population(pop, @zdt2); % Use @zdt2 instead of @zdt1 for gen = 1:T % Non-dominated sorting [fronts, ranks] = non_dominated_sort(fitness); % Create mating pool using tournament selection mating_pool = tournament_selection(pop, ranks, NPop); % Generate offspring using crossover and mutation offspring = crossover(mating_pool, 0.9, dim); offspring = mutation(offspring, 1/dim, dim); % Evaluate offspring offspring_fitness = evaluate_population(offspring, @zdt2); % Use @zdt2 % Combine population and offspring comb_pop = [pop; offspring]; comb_fitness = [fitness; offspring_fitness]; % Non-dominated sorting on combined population [fronts, ~] = non_dominated_sort(comb_fitness); % Create new population [pop, fitness] = select_new_population(comb_pop, comb_fitness, fronts, NPop); % Display generation progress disp(['Generation #', num2str(gen), ' completed.']); end % Return last population as Pareto front paretoFront = fitness; paretoSet = pop; end % Add your NSGA-II helper functions here (initialization, evaluate_population, non_dominated_sort, dominates, tournament_selection, crossover, mutation, select_new_population, and crowding_distance).Both scripts now use the ZDT2 problem, which should allow you to compare the performance of MOEMA and NSGA-II. Make sure to fill in the helper functions for both scripts according to the already-provided implementations, adjusting as necessary to ensure both variants are solving the ZDT2 function.
- User
combine and compare it in one script
- Assistant
To combine both the MOEMA and NSGA-II algorithms into a single script for comparison on the ZDT2 problem, we can organize the code to run one after the other and plot their results together. Below is a combined MATLAB script:
clc; clear; close all; % Set problem parameters NPop = 100; % Population size T = 200; % Number of iterations dim = 30; % Number of decision variables lb = zeros(1, dim); % Lower bound ub = ones(1, dim); % Upper bound % Run MOEMA for ZDT2 [paretoFrontMOEMA, paretoSetMOEMA] = MOEMA(NPop, T, lb, ub, dim, @zdt2); % Run NSGA-II for ZDT2 [paretoFrontNSGA2, paretoSetNSGA2] = NSGA2(NPop, T, lb, ub, dim); % Plot comparison of Pareto fronts figure; hold on; plot(paretoFrontMOEMA(:,1), paretoFrontMOEMA(:,2), 'ro', 'DisplayName', 'MOEMA'); plot(paretoFrontNSGA2(:,1), paretoFrontNSGA2(:,2), 'bo', 'DisplayName', 'NSGA-II'); xlabel('f1'); ylabel('f2'); title('Comparison of MOEMA and NSGA-II on ZDT2'); legend('show'); grid on; hold off; % MOEMA Function function [paretoFront, paretoSet] = MOEMA(NPop, T, lb, ub, dim, FITNESSFCN) fitness_history = zeros(NPop, T, 2); % Assuming 2 objectives position_history = zeros(NPop, T, dim); pop = initialization(NPop, dim, ub, lb); pop_dad = pop(1:NPop/2, :); pop_mum = pop(NPop/2+1:end, :); obj_dad = evaluate_population(pop_dad, FITNESSFCN); obj_mum = evaluate_population(pop_mum, FITNESSFCN); obj = [obj_dad; obj_mum]; archive = update_archive([], [], pop, obj); for gen = 1:T k1 = randperm(NPop/2); k2 = randperm(NPop/2); k1x = pop_dad(k1, :); k2x = pop_mum(k2, :); for k = 1:NPop/2 Imates = 1 + (var(k1x(k, :)) - var(k2x(k, :))); p = randn(1, dim); r = rand(); if Imates >= 0 temp1 = p .* k1x(k, :); temp2 = (1-p) .* k2x(k, :); pop_child = temp1 + temp2; else temp1 = p .* k2x(k, :); temp2 = (1-p) .* k1x(k, :); pop_child = temp1 + temp2; end if ~isempty(archive.pop) best_pop = archive.pop(randi(size(archive.pop, 1)), :); K = rand(1, dim) < 0.8; pop_child = best_pop .* (1-K) + pop_child .* K; end pop_child = max(min(ub, pop_child), lb); obj_child = FITNESSFCN(pop_child); [pop_dad, obj_dad] = update_population(pop_dad, obj_dad, pop_child, obj_child, k1(k)); [pop_mum, obj_mum] = update_population(pop_mum, obj_mum, pop_child, obj_child, k2(k)); if r < 0.2 && ~isempty(archive.pop) temp1 = archive.pop(randi(size(archive.pop, 1)), :); pop_child = rand(1, dim) .* temp1; pop_child = max(min(ub, pop_child), lb); obj_child = FITNESSFCN(pop_child); [pop_dad, obj_dad] = update_population(pop_dad, obj_dad, pop_child, obj_child, randi(NPop/2)); [pop_mum, obj_mum] = update_population(pop_mum, obj_mum, pop_child, obj_child, randi(NPop/2)); end end pop = [pop_dad; pop_mum]; obj = [obj_dad; obj_mum]; archive = update_archive(archive.pop, archive.obj, pop, obj); fitness_history(:, gen, :) = obj; position_history(:, gen, :) = pop; end paretoFront = archive.obj; paretoSet = archive.pop; end % NSGA-II Function function [paretoFront, paretoSet] = NSGA2(NPop, T, lb, ub, dim) pop = initialization(NPop, dim, ub, lb); fitness = evaluate_population(pop, @zdt2); % Use @zdt2 for gen = 1:T [fronts, ranks] = non_dominated_sort(fitness); mating_pool = tournament_selection(pop, ranks, NPop); offspring = crossover(mating_pool, 0.9, dim); offspring = mutation(offspring, 1/dim, dim); offspring_fitness = evaluate_population(offspring, @zdt2); comb_pop = [pop; offspring]; comb_fitness = [fitness; offspring_fitness]; [fronts, ~] = non_dominated_sort(comb_fitness); [pop, fitness] = select_new_population(comb_pop, comb_fitness, fronts, NPop); end paretoFront = fitness; paretoSet = pop; end % Common Functions function pop = initialization(NPop, dim, ub, lb) pop = rand(NPop, dim) .* (ub - lb) + lb; end function obj = evaluate_population(pop, FITNESSFCN) obj = zeros(size(pop, 1), 2); for i = 1:size(pop, 1) obj(i, :) = FITNESSFCN(pop(i, :)); end end function [pop, obj] = update_population(pop, obj, child, child_obj, index) if dominates(child_obj, obj(index, :)) pop(index, :) = child; obj(index, :) = child_obj; elseif ~dominates(obj(index, :), child_obj) if rand() < 0.5 pop(index, :) = child; obj(index, :) = child_obj; end end end function archive = update_archive(archive_pop, archive_obj, pop, obj) combined_pop = [archive_pop; pop]; combined_obj = [archive_obj; obj]; [is_dominated, ~] = pareto_dominance(combined_obj); non_dominated = ~is_dominated; archive.pop = combined_pop(non_dominated, :); archive.obj = combined_obj(non_dominated, :); end function [is_dominated, domination_count] = pareto_dominance(objectives) [n_solutions, n_objectives] = size(objectives); is_dominated = false(n_solutions, 1); domination_count = zeros(n_solutions, 1); for i = 1:n_solutions for j = i+1:n_solutions if dominates(objectives(i, :), objectives(j, :)) is_dominated(j) = true; domination_count(i) = domination_count(i) + 1; elseif dominates(objectives(j, :), objectives(i, :)) is_dominated(i) = true; domination_count(j) = domination_count(j) + 1; end end end end function result = dominates(obj1, obj2) result = all(obj1 <= obj2) && any(obj1 < obj2); end function [fronts, rank] = non_dominated_sort(fitness) num_individuals = size(fitness, 1); domination_count = zeros(num_individuals, 1); dominated_set = cell(num_individuals, 1); rank = zeros(num_individuals, 1); fronts = cell(1); for i = 1:num_individuals for j = 1:num_individuals if i ~= j if dominates(fitness(i, :), fitness(j, :)) dominated_set{i} = [dominated_set{i}, j]; elseif dominates(fitness(j, :), fitness(i, :)) domination_count(i) = domination_count(i) + 1; end end end if domination_count(i) == 0 rank(i) = 1; fronts{1} = [fronts{1}, i]; end end current_front = 1; while ~isempty(fronts{current_front}) next_front = []; for i = fronts{current_front} for j = dominated_set{i} domination_count(j) = domination_count(j) - 1; if domination_count(j) == 0 rank(j) = current_front + 1; next_front = [next_front, j]; end end end current_front = current_front + 1; fronts{current_front} = next_front; end fronts = fronts(1:current_front-1); end function selected_pop = tournament_selection(pop, ranks, num_selected) selected_pop = zeros(num_selected, size(pop, 2)); for i = 1:num_selected candidates = randperm(size(pop, 1), 2); if ranks(candidates(1)) < ranks(candidates(2)) selected_pop(i, :) = pop(candidates(1), :); else selected_pop(i, :) = pop(candidates(2), :); end end end function offspring = crossover(pop, crossover_prob, num_vars) offspring = zeros(size(pop)); for i = 1:2:size(pop, 1) if rand < crossover_prob parent1 = pop(i, :); parent2 = pop(i+1, :); cross_point = randi([1, num_vars-1], 1, 1); offspring(i, :) = [parent1(1:cross_point), parent2(cross_point+1:end)]; offspring(i+1, :) = [parent2(1:cross_point), parent1(cross_point+1:end)]; else offspring(i, :) = pop(i, :); offspring(i+1, :) = pop(i+1, :); end end end function mutated_pop = mutation(pop, mutation_prob, num_vars) eta_m = 20; mutated_pop = pop; for i = 1:size(pop, 1) for j = 1:num_vars if rand < mutation_prob r = rand; if r < 0.5 delta = (2 * r)^(1/(eta_m + 1)) - 1; else delta = 1 - (2 * (1 - r))^(1/(eta_m + 1)); end mutated_pop(i, j) = pop(i, j) + delta; mutated_pop(i, j) = min(max(mutated_pop(i, j), 0), 1); end end end end function [pop, fitness] = select_new_population(comb_pop, comb_fitness, fronts, NPop) pop = []; fitness = []; for i = 1:length(fronts) if size(pop, 1) + length(fronts{i}) > NPop break; end pop = [pop; comb_pop(fronts{i}, :)]; fitness = [fitness; comb_fitness(fronts{i}, :)]; end remaining = NPop - size(pop, 1); if remaining > 0 selected_front = fronts{i}; distances = crowding_distance(comb_fitness(selected_front, :)); [~, sorted_indices] = sort(distances, 'descend'); selected = selected_front(sorted_indices(1:remaining)); pop = [pop; comb_pop(selected, :)]; fitness = [fitness; comb_fitness(selected, :)]; end end function distances = crowding_distance(fitness) num_individuals = size(fitness, 1); num_objectives = size(fitness, 2); distances = zeros(num_individuals, 1); for m = 1:num_objectives [sorted_values, sorted_idx] = sort(fitness(:, m)); distances(sorted_idx(1)) = inf; distances(sorted_idx(end)) = inf; for i = 2:num_individuals-1 if sorted_values(end) ~= sorted_values(1) distances(sorted_idx(i)) = distances(sorted_idx(i)) + ... (sorted_values(i+1) - sorted_values(i-1)) / (sorted_values(end) - sorted_values(1)); end end end end function f = zdt2(x) f1 = x(1); g = 1 + 9*sum(x(2:end))/(length(x)-1); f2 = g * (1 - (f1/g)^2); f = [f1, f2]; endExplanation:
-
Initialization: Both algorithms are initialized with the same problem parameters (
NPop,T,dim,lb,ub). -
MOEMA and NSGA-II: Both algorithms are executed sequentially, using the ZDT2 function as the fitness function.
-
Plotting: The script plots the final Pareto fronts obtained from MOEMA and NSGA-II in the same figure to visually compare their performance.
-
Helper Functions: Common helper functions for initialization, fitness evaluation, domination checks, sorting, selection, crossover, mutation, and computation of the ZDT2 objectives are included. These functions are shared between the two algorithms.
This setup allows for direct comparison of MOEMA and NSGA-II on the ZDT2 multi-objective problem, evaluating their ability to find diverse and optimal Pareto fronts. Adjust the plotting and metrics as necessary for your specific analysis requirements.
-