rheuristic2xover.sci

来自「基于SCILAB的The Genetic Algorithm Toolbox f」· SCI 代码 · 共 38 行

function [c1,c2] = RHeuristic2Xover(p1,p2,bounds,Ops)
// Heuristic crossover takes two parents P1,P2 and performs an extrapolation
// along the line formed by the two parents outward in the direction of the
// better parent.
//
// function [c1,c2] = heuristicXover(p1,p2,bounds,Ops)
// p1      - the first parent ( [solution string function value] )
// p2      - the second parent ( [solution string function value] )
// bounds  - the bounds matrix for the solution space
// Ops     - Options for heuristic crossover, [gen #heurXovers number_of_retries]

b1=bounds(:,1)';
b2=bounds(:,2)';
numVar = size(p1,2)-1;
// Determine the best and worst parent
if(p1(numVar+1) > p2(numVar+1))
  bt = p1; 
  wt = p2;
else
  bt = p2;
  wt = p1;
end
// Pick a random mix amount
a = rand;
// Create the child
c1 = a * (bt - wt) + bt;
  
// Check to see if child is within bounds
outrange= (c1(1:numVar) > b2 )
if any(outrange)
  c1(find(outrange>0))=b2(find(outrange>0));
end

// Crossover functions return two children therefore return the best
// and the new child created
c2 = p1*p1(numVar+1)/(p2(numVar+1)+p1(numVar+1)) + p2*p2(numVar+1)/(p2(numVar+1)+p1(numVar+1));