nonunifmutation.m

来自「18个遗传算法程序matlab源码」· M 代码 · 共 28 行

function [parent] = nonUnifMutate(parent,bounds,Ops)
% Non uniform mutation changes one of the parameters of the parent
% based on a non-uniform probability distribution.  This Gaussian
% distribution starts wide, and narrows to a point distribution as the
% current generation approaches the maximum generation.
%
% function [newSol] = nonUnifMutate(parent,bounds,Ops)
% parent  - the first parent ( [solution string function value] )
% bounds  - the bounds matrix for the solution space
% Ops     - Options for nonUnifMutate[gen #nonUnifMutations maxGen b]


cg=Ops(1); 			         	% Current Generation
mg=Ops(3);                    % Maximum Number of Generations
b=Ops(4);                     % Shape parameter
df = bounds(:,2) - bounds(:,1); 	% Range of the variables
numVar = size(parent,2)-1; 		% Get the number of variables
% Pick a variable to mutate randomly from 1 to number of vars
mPoint = round(rand * (numVar-1)) + 1;
md = round(rand); 			% Choose a direction of mutation
if md 					% Mutate towards upper bound
  newValue=parent(mPoint)+delta(cg,mg,bounds(mPoint,2)-parent(mPoint),b);
else 					% Mutate towards lower bound
  newValue=parent(mPoint)-delta(cg,mg,parent(mPoint)-bounds(mPoint,1),b);
end
parent(mPoint) = newValue; 		% Make the child