相量粒子群优化算法附matlab代码

% Based on the paper: % Ghasemi, Mojtaba, Ebrahim Akbari, Abolfazl Rahimnejad, % Seyed Ehsan Razavi, Sahand Ghavidel, and Li Li. % "Phasor particle swarm optimization: a simple and efficient % variant of PSO." Soft Computing 23, no. 19 (2019): 9701-9718. clear clc disp('CLPPSO'); CostFunction=@(x) Cost(x); nvar=30; D=nvar; xmin=-100; xmax=-xmin; dx=xmax-xmin; vmax=0.5*dx; npop=40; maxit=100000/npop; ai = zeros(npop, nvar); f_pbest = 1 : npop; f_pbest = repmat(f_pbest', 1, nvar); t = 0 : 1 / (npop - 1) : 1; t = 5 .* t; Pc = 0.0 + (0.5 - 0.0) .* (exp(t) - exp(t(1))) ./ (exp(t(npop)) - exp(t(1))); m = 0 .* ones(npop, 1); for it=1:maxit if it==1 gbestcost(1)=inf; for i=1:npop velocity(i,:)=zeros(1,nvar); delta(i)=unifrnd(0,2*pi); position(i,:)=xmin+(xmax-xmin)*rand(1,nvar); cost(i)=CostFunction(position(i,:));%%,M ,o ,A, a, b, A MM(i)=7; FF(i)=0; pbest(i,:)=position(i,:); pbestcost(i)=cost(i); if pbestcost(i)<gbestcost(it) gbest=pbest(i,1:nvar); gbestcost(it)=pbestcost(i); end end for k = 1 : npop ar = randperm(nvar); ai(k, ar(1 : m(k))) = 1; fi1 = ceil(npop * rand(1, nvar)); fi2 = ceil(npop * rand(1, nvar)); fi = (pbestcost(fi1) < pbestcost(fi2)) .* fi1 + (pbestcost(fi1) >= pbestcost(fi2)) .* fi2; bi = ceil(rand(1, nvar) - 1 + Pc(k)); if bi == zeros(1, nvar), rc = randperm(nvar); bi(rc(1)) = 1; end f_pbest(k, :) = bi .* fi + (1 - bi) .* f_pbest(k, :); end else gbestcost(it)=gbestcost(it-1); for i=1:npop if FF(i)>=2 FF(i)=0; ai(i, :) = zeros(1, nvar); f_pbest(i, :) = i .* ones(1,nvar); ar = randperm(nvar); ai(i, ar(1 : m(i))) = 1; fi1 = ceil(npop * rand(1, nvar)); fi2 = ceil(npop * rand(1, nvar)); fi = (pbestcost(fi1) < pbestcost(fi2)) .* fi1 + (pbestcost(fi1) >= pbestcost(fi2)) .* fi2; bi = ceil(rand(1, nvar) - 1 + Pc(i)); if bi == zeros(1, nvar), rc = randperm(nvar); bi(rc(1)) = 1; end f_pbest(i, :) = bi .* fi + (1 - bi) .* f_pbest(i, :); end aa=2*(sin(delta(i))); bb=2*(cos(delta(i))); ee=abs(cos(delta(i)))^aa; tt=abs(sin(delta(i)))^bb; for j1=1:nvar velocity(i,j1)=((ee/(npop))*velocity(i,j1))+(ee)*(pbest(f_pbest(i, j1),j1)-position(i,j1)); end velocity(i,:)=min(max(velocity(i,:),-vmax),vmax); position(i,:)=position(i,:)+velocity(i,:); position(i,:)=min(max(position(i,:),xmin),xmax); cost(i)=CostFunction(position(i,:)); % delta(i)=delta(i)+(abs(aa+bb)*(2*pi)); vmax=(abs(cos(delta(i)))^2)*dx; if cost(i)<pbestcost(i) FF(i)=0; else FF(i)=FF(i)+1; end if cost(i)<pbestcost(i) pbest(i,:)=position(i,:); pbestcost(i)=cost(i); else if pbestcost(i)<gbestcost(it) gbest=pbest(i,:); gbestcost(it)=pbestcost(i); end end end end disp(['Iteration ' num2str(it) ': Best Cost = ' num2str(gbestcost(it))]); end hold on; figure(1) it=1:50:maxit; semilogy((it),(log(gbestcost(it))),'-or','linewidth',1.4,'Markersize',3); xlabel('\fontsize{12}\bf Iteration'); ylabel('\fontsize{12}\bf Best value'); legend('\fontsize{10}\bf PPSO',1); Cost_Rsult=gbestcost(end) %% toc