GA.rar_GA十进制算法_fullyofp_matlab混合遗传_混合遗传算法_遗传算法

clear; %--------------------------------- % 初始参数 num_pop = 200; %种群中的个体数量 hy_prob = 0.7; %种群的杂交率 mu_prob = 0.2; %个体变异率 prec = 10^-10; %可接受的精度要求 genetic_round = 1000; %遗传的最大代数 genetic_round0 = 1; %遗传的最小代数 num_bestkeeper = 20; %保优数量 eps_obj = 10^-10; eps_var = 10^-10; eta = 1; permutation = 1; %--------------------------------- % 设计变量阈值 %lb = -5; %ub=5; lb = [5;5]; %设计变量下限值，列向量 ub = [10;10]; %设计变量上限值，列向量 %--------------------------------- % 目标函数句柄 obj = @ObjFun; index_min_max = -1; %最大化或者最小化 %--------------------------------- % 十进制编码遗传算法 x_op = zeros(length(ub),genetic_round); %初始化最优点存储矩阵 obj_op = zeros(1,genetic_round); %初始化最优目标值向量 len_code = CodeLen( lb,ub,prec ); %计算编码长度 population = IniPopulGen( num_pop,lb,ub ); %生成初始种群 for kk = 1:1:genetic_round [ fitness01,cumulation_fitness ] = Fitness( population,obj,index_min_max ); %计算累积适应度分布 population = Selection( population,fitness01,cumulation_fitness,num_bestkeeper ); %选择 %population = SelectionLamarc( population,fitness01,eta,permutation); %获得性遗传筛选 code_dec = CodingReal2Dec( population,lb,ub,len_code ); %十进制编码 code_bin = CodingDec2Bin( code_dec,len_code ); %二次编码，二进制编码 code_hy = HybridBin( code_bin,hy_prob,num_pop,len_code ); %杂交 code_mu = MutationBin( code_hy,mu_prob,num_pop,len_code); %变异 code_dec = DecodingBin2Dec( code_mu,len_code ); %第一层解码 code_dec0 = code_dec; population = DecodingDec2Real( code_dec,lb,ub,len_code ); %解码 obj_val = ObjFun(population); %计算目标值 [obj_optemp,x_op_ind] = min(obj_val); %当前最优解 obj_op(kk) = obj_optemp; x_op(:,kk) = population(:,x_op_ind); if kk > genetic_round0 if (abs(obj_op(kk)-obj_op(kk-1))<eps_var) break; end end disp('遗传代数：'); disp(kk); end disp('最优目标值：'); disp(obj_op(kk)); disp('最优变量值：'); disp(x_op(:,kk)); %---------------------------------- % 目标函数 function object_val = ObjFun(x) %object_val = (x(1,:)-pi).^2+(x(2,:)-exp(1)).^2-3; %object_val = (x-exp(1)).^2-3; %object_val = x.^2+x+5*sin(3*x)+3*cos(4*x); object_val = x(1,:).*sin(4*x(1,:))+1.1*x(2,:).*sin(2*x(2,:)); end