基于遗传算法的模糊推理优化MFs

时间:2014-01-06 12:26:37

标签: matlab optimization genetic-algorithm fuzzy-logic

我正在使用 GA 来优化模糊系统中隶属函数的参数。 我创建了一个健身功能:

function y = gafuzzy(x)
global FISsys
global allData
global realResult

FISsys = readfis('aCAess.fis'); 
allData = importdata('ab.mat');
realResult = importdata('ad.mat');

FISsys.input(1,1).mf(1,1).params = [x(1) x(2) x(3)];
FISsys.input(1,1).mf(1,2).params = [x(4) x(5) x(6)];

FISsys.input(1,2).mf(1,1).params = [x(7) x(8) x(9)];
FISsys.input(1,2).mf(1,2).params = [x(10) x(11) x(12)];

FISsys.output.mf(1,1).params = [x(13) x(14) x(15)];
FISsys.output.mf(1,2).params = [x(16) x(17) x(18)];

c = evalfis(allData,FISsys);
e=sum(abs(c-realResult)); 
y = e;

end

线性不等式的[15 * 18]矩阵是:

A = [1,-1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0;
0,1,-1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0;
0,0,0,1,-1,0,0,0,0,0,0,0,0,0,0,0,0,0;
0,0,0,0,1,-1,0,0,0,0,0,0,0,0,0,0,0,0;
0,0,0,0,0,0,1,-1,0,0,0,0,0,0,0,0,0,0;
0,0,0,0,0,0,0,1,-1,0,0,0,0,0,0,0,0,0;
0,0,0,0,0,0,0,0,0,1,-1,0,0,0,0,0,0,0;
0,0,0,0,0,0,0,0,0,0,1,-1,0,0,0,0,0,0;
0,0,0,0,0,0,0,0,0,0,0,0,1,-1,0,0,0,0;
0,0,0,0,0,0,0,0,0,0,0,0,0,1,-1,0,0,0;
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,-1,0;
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,-1;
0,1,0,0,-1,0,0,0,0,0,0,0,0,0,0,0,0,0;
0,0,0,0,0,0,0,1,0,0,-1,0,0,0,0,0,0,0;
0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,-1,0]

和b [15 * 1]向量是:

b = [0;0;0;0;0;0;0;0;0;0;0;0;0;0;0]

但是当我运行 GA 时,我收到此错误:     fisTriangleMf()中的非法参数 - > a> B'/ P>

为什么?

1 个答案:

答案 0 :(得分:3)

通常,在三角形MF中,第一个数字,这里a(显示左顶点)应该小于第二个数字,这里是b(顶部顶点)。所以你可以有一个像[-1 0 1]这样的三角形MF,但它不能像[0 -1 1]。

在你的代码中,我认为有时你不满足其中一个地方的不平等:

[x(1) < x(2) < x(3)];
[x(4) < x(5) < x(6)];
[x(7) < x(8) < x(9)];
[x(10) < x(11) < x(12)];
....

如果程序随机化这些值,您可以通过检查和替换轻松地将它们绑定到代码中,例如:

if x(1) >= x(2)
     tmp = x(1);
     x(1) = x(2);
     x(2) = tmp;
end