【WSN覆盖优化】基于改进黏菌算法的无线传感器网络覆盖 WSN覆盖优化【matlab代码】

function [Destination_fitness,bestPositionsX,bestPositionsY,Convergence_curve]=ISMA(N,Max_iter,lb,ub,dim,objfun,R,w) disp('SMA is now tackling your problem') % initialize position bestPositionsX=zeros(1,dim); bestPositionsY=zeros(1,dim); Destination_fitness=-inf;%change this to -inf for maximization problems AllFitness = -inf*ones(N,1);%record the fitness of all slime mold weightX = ones(N,dim);%fitness weight of each slime mold weightY = ones(N,dim);%fitness weight of each slime mold %Initialize the set of random solutions % X=initialization(N,dim,ub,lb); for i = 1:N PosX(i,:)=rand(1,dim).*(ub-lb)+lb; PosY(i,:)=rand(1,dim).*(ub-lb)+lb; end Convergence_curve=zeros(1,Max_iter); it=1; %Number of iterations % lb=ones(1,dim).*lb; % lower boundary % ub=ones(1,dim).*ub; % upper boundary z=0.03; % parameter % Main loop while it <= Max_iter display(['当前迭代次数为 : ', num2str(it)]); %sort the fitness for i=1:N % Check if solutions go outside the search space and bring them back Flag4ubX=PosX(i,:)>ub; Flag4lbX=PosX(i,:)<lb; PosX(i,:)=(PosX(i,:).*(~(Flag4ubX+Flag4lbX)))+ub.*Flag4ubX+lb.*Flag4lbX; Flag4ubY=PosY(i,:)>ub; Flag4lbY=PosY(i,:)<lb; PosY(i,:)=(PosY(i,:).*(~(Flag4ubY+Flag4lbY)))+ub.*Flag4ubY+lb.*Flag4lbY; AllFitness(i) = feval(objfun,PosX(i,:),PosY(i,:),R,w); end [SmellOrder,SmellIndex] = sort(AllFitness,'descend'); %Eq.(2.6) worstFitness = SmellOrder(N); bestFitness = SmellOrder(1); S=bestFitness-worstFitness+eps; % plus eps to avoid denominator zero %calculate the fitness weight of each slime mold for i=1:N for j=1:dim if i<=(N/2) %Eq.(2.5) weightX(SmellIndex(i),j) = 1+rand()*log10((bestFitness-SmellOrder(i))/(S)+1); weightY(SmellIndex(i),j) = 1+rand()*log10((bestFitness-SmellOrder(i))/(S)+1); else weightX(SmellIndex(i),j) = 1-rand()*log10((bestFitness-SmellOrder(i))/(S)+1); weightY(SmellIndex(i),j) = 1-rand()*log10((bestFitness-SmellOrder(i))/(S)+1); end end end %update the best fitness value and best position if bestFitness > Destination_fitness bestPositionsX=PosX(SmellIndex(1),:); bestPositionsY=PosY(SmellIndex(1),:); Destination_fitness = bestFitness; end a = atanh(-(it/Max_iter)+1); %Eq.(2.4) b = 1-it/Max_iter; % Update the Position of search agents for i=1:N if rand<z %Eq.(2.7) PosX(i,:) = rand(1,dim).*(ub-lb)+lb; PosY(i,:) = rand(1,dim).*(ub-lb)+lb; else % p =tanh(abs(AllFitness(i)-Destination_fitness)); %Eq.(2.2) % ★★改进1：参数P的改进★★ p = SmellIndex(i)/N; vbX = unifrnd(-a,a,1,dim); %Eq.(2.3) vbY = unifrnd(-a,a,1,dim); vcX = unifrnd(-b,b,1,dim); vcY = unifrnd(-b,b,1,dim); for j=1:dim r = rand(); AX = randi([1,N]); % two positions randomly selected from population AY = randi([1,N]); BX = randi([1,N]); BY = randi([1,N]); if r<p %Eq.(2.1) PosX(i,j) = bestPositionsX(j)+ vbX(j)*(weightX(i,j)*PosX(AX,j)-PosX(BX,j)); PosY(i,j) = bestPositionsY(j)+ vbY(j)*(weightY(i,j)*PosY(AY,j)-PosY(BY,j)); else PosX(i,j) = vcX(j)*PosX(i,j); PosY(i,j) = vcY(j)*PosY(i,j); end end end end % ★★改进2：混沌精英突变★★ for j = 1:dim faiX(1,j)=(bestPositionsX(j)-lb(j))/(ub(j)-lb(j)); newfaiX(1,j)=mod(faiX(1,j)+0.2-(0.25./pi),1); faiY(1,j)=(bestPositionsY(j)-lb(j))/(ub(j)-lb(j)); newfaiY(1,j)=mod(faiY(1,j)+0.2-(0.25./pi),1); end H=1-(it/Max_iter); for j = 1:dim newbestPositionsX(1,j)=H*bestPositionsX(j)+(1-H)*bestPositionsX(j)*newfaiX(1,j); newbestPositionsY(1,j)=H*bestPositionsY(j)+(1-H)*bestPositionsY(j)*newfaiY(1,j); end tempFit = feval(objfun,newbestPositionsX,newbestPositionsY,R,w); if tempFit>feval(objfun,bestPositionsX,bestPositionsY,R,w); bestPositionsX = newbestPositionsX; bestPositionsY = newbestPositionsY; Destination_fitness = tempFit; end Convergence_curve(it)=Destination_fitness; it=it+1; end end