合作协同进化（Cooperative Coevolution）是求解大规模优化算法一个有效的方法。将大规模问题分解为一组组较小的子问题。而合作协同进化的关键是分解策略。
合作协同进化算法请见：https://www.omegaxyz.com/2017/10/14/cooperative_coevolution/
NSGA2算法是一种多目标遗传算法。此文章是随机固定分组的合作协同进化利用NSGA2来优化。
比如有12个决策变量，我们固定随机优化3个决策变量，那么就将决策变量分成了4组。
MATLAB主函数代码：
MATLAB
代码语言：txt复制%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
clc;
clear;
global pop
pop = 500; %种群数量
gen = 2; %迭代次数
global M
M = 2; %目标数量
Dim=22; %搜索空间维数（未知数个数）
sub_dim= 2 ;
global min_range
global max_range
min_range = zeros(1, Dim); %下界
max_range = ones(1,Dim); %上界
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
divide_datasets();
global answer
answer=cell(M,3);
Dim_index = ones(1,1)*(1:Dim 4);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
chromosome = initialize_variables(pop, M, Dim, min_range, max_range, Dim_index);
chromosome = non_domination_sort_mod(chromosome, M, Dim);
result = 1;
while gen ~= 0
subgroup = rnd_divide(Dim, sub_dim);
for i=1:length(subgroup)
subgroup{i}(sub_dim 1)=Dim 1;
subgroup{i}(sub_dim 2)=Dim 2;
subgroup{i}(sub_dim 3)=Dim 3;
subgroup{i}(sub_dim 4)=Dim 4;
[temp_chromosome] = nsga2(chromosome(:,subgroup{i}), sub_dim, subgroup{i});
chromosome(:,subgroup{i}(1:sub_dim)) = temp_chromosome(:,1:sub_dim);
end
chromosome = nsga2(chromosome, Dim, Dim_index);
chromosome = non_domination_sort_mod(chromosome, M, Dim);
gen =gen - 1;
progress = 1-gen/10
end
plot(chromosome(:,Dim 1),chromosome(:,Dim 2),'*');
xlabel('f_1'); ylabel('f_2');
title('Pareto Optimal Front');随机分组代码：
MATLAB
代码语言：txt复制% random grouping
function group = rnd_divide(dim, subdim)
dim_rand = randperm(dim);
group = {};
for i = 1:subdim:dim
index = dim_rand(i:i subdim-1);
group = {group{1:end} index};
end
end其它的算法依赖函数与自定义优化函数的NSGA2算法相同https://www.omegaxyz.com/2018/01/22/new_nsga2/
