fitness.m

该算法是综合遗传算法和神经网络算法的综合性算法

function fitnessval=fitness(chrom,p,aim)
global   P_cross  P_mutation sumen
load training_data1; 
[Popsize len]=size(chrom);
fitness_gene=zeros(Popsize,1);
in_he=zeros(2,1);
out_he=zeros(1);
in_out=0;
out_out=0;
% 权值和阈值的分配设置
for i=1:Popsize
    w_he=[chrom(i,1);
          chrom(i,2);];
%     w_out=[chrom(i,5)];
    b_he=chrom(i,3);
    a=chrom(i,4);
    
    in_he=w_he(1)*p(1)+w_he(2)*p(2)+b_he;
    out_he=(1-exp(-a*in_he))/(1+exp(a*in_he));
    
%     error=zeros(2,1);
%     for j=1:2
%         for s=1:2
%             % 隐层的输入与输出    
%             in_he(s)=sum(w_he(:,s).*p(:,j))+b_he(s);%隐层的输入
% %             if in_he(s)>1.0e002
% %                 out_he(s)=-1;
% %             else
%                 out_he(s)=(1-exp(in_he(s)))/(1+exp(in_he(s)));      %隐层的输出
% %             end
%         end
%         % 输出层的输入与输出
%         in_out=w_out*out_he+b_out;
%         out_out=1/(1+exp(-in_out)); 
%         
%         error(j)=aim(j)-out_out;
%     end
%     fitness_gene(i)=8/sum(error.^2); % 染色体均方误差
    
end    
for in=1:n
        m=(jn-1)*n+in+1
        k1=fun5(x,F,zF)+un;
        k2=fun5(x+h*k1/2,F,zF)+un;
        k3=fun5(x+h*k2/2,F,zF)+un;
        k4=fun5(x+h*k3,F,zF)+un;
        x=x+h/6*(k1+2*k2+2*k3+k4); 
        yn=x(:,1);
        y1(m)=x(:,41);
        if yn<0
            error('yn<0请重新设置控制参数')
        end
        y(m)=yn;
        t1(m)=(m-1)*h;
end   
    en=r-yn; 
    e(jn+1)=en;
    sumen=sumen+abs(e(jn+1));

save data1;
    fitnessval=sumen;
% y=fitness_gene;