杂草算法（Invasive Weed Optimization， IWO）

Halcom

杂草算法（Invasive Weed Optimization， IWO）

1. % Developer: S. Mostapha Kalami Heris (Member of Yarpiz Team)
   
2. %
   
3. % Contact Info: sm.kalami@gmail.com, info@yarpiz.com
   
4. %
   
5. 
   
6. clc;
   
7. clear;
   
8. close all;
   
9. 
   
10. %% Problem Definition
    
11. 
    
12. CostFunction = @(x) Sphere(x);  % Objective Function
    
13. 
    
14. nVar = 5;           % Number of Decision Variables
    
15. VarSize = [1 nVar]; % Decision Variables Matrix Size
    
16. 
    
17. VarMin = -10;       % Lower Bound of Decision Variables
    
18. VarMax = 10;        % Upper Bound of Decision Variables
    
19. 
    
20. %% IWO Parameters
    
21. 
    
22. MaxIt = 200;    % Maximum Number of Iterations
    
23. 
    
24. nPop0 = 10;     % Initial Population Size
    
25. nPop = 25;      % Maximum Population Size
    
26. 
    
27. Smin = 0;       % Minimum Number of Seeds
    
28. Smax = 5;       % Maximum Number of Seeds
    
29. 
    
30. Exponent = 2;           % Variance Reduction Exponent
    
31. sigma_initial = 0.5;    % Initial Value of Standard Deviation
    
32. sigma_final = 0.001;        % Final Value of Standard Deviation
    
33. 
    
34. %% Initialization
    
35. 
    
36. % Empty Plant Structure
    
37. empty_plant.Position = [];
    
38. empty_plant.Cost = [];
    
39. 
    
40. pop = repmat(empty_plant, nPop0, 1);    % Initial Population Array
    
41. 
    
42. for i = 1:numel(pop)
    
43.    
    
44.     % Initialize Position
    
45.     pop(i).Position = unifrnd(VarMin, VarMax, VarSize);
    
46.    
    
47.     % Evaluation
    
48.     pop(i).Cost = CostFunction(pop(i).Position);
    
49.    
    
50. end
    
51. 
    
52. % Initialize Best Cost History
    
53. BestCosts = zeros(MaxIt, 1);
    
54. 
    
55. %% IWO Main Loop
    
56. 
    
57. for it = 1:MaxIt
    
58.    
    
59.     % Update Standard Deviation
    
60.     sigma = ((MaxIt - it)/(MaxIt - 1))^Exponent * (sigma_initial - sigma_final) + sigma_final;
    
61.    
    
62.     % Get Best and Worst Cost Values
    
63.     Costs = [pop.Cost];
    
64.     BestCost = min(Costs);
    
65.     WorstCost = max(Costs);
    
66.    
    
67.     % Initialize Offsprings Population
    
68.     newpop = [];
    
69.    
    
70.     % Reproduction
    
71.     for i = 1:numel(pop)
    
72.         
    
73.         ratio = (pop(i).Cost - WorstCost)/(BestCost - WorstCost);
    
74.         S = floor(Smin + (Smax - Smin)*ratio);
    
75.         
    
76.         for j = 1:S
    
77.             
    
78.             % Initialize Offspring
    
79.             newsol = empty_plant;
    
80.             
    
81.             % Generate Random Location
    
82.             newsol.Position = pop(i).Position + sigma * randn(VarSize);
    
83.             
    
84.             % Apply Lower/Upper Bounds
    
85.             newsol.Position = max(newsol.Position, VarMin);
    
86.             newsol.Position = min(newsol.Position, VarMax);
    
87.             
    
88.             % Evaluate Offsring
    
89.             newsol.Cost = CostFunction(newsol.Position);
    
90.             
    
91.             % Add Offpsring to the Population
    
92.             newpop = [newpop
    
93.                       newsol];  %#ok
    
94.             
    
95.         end
    
96.         
    
97.     end
    
98.    
    
99.     % Merge Populations
    
100.     pop = [pop
     
101.            newpop];
     
102.    
     
103.     % Sort Population
     
104.     [~, SortOrder]=sort([pop.Cost]);
     
105.     pop = pop(SortOrder);
     
106. 
     
107.     % Competitive Exclusion (Delete Extra Members)
     
108.     if numel(pop)>nPop
     
109.         pop = pop(1:nPop);
     
110.     end
     
111.    
     
112.     % Store Best Solution Ever Found
     
113.     BestSol = pop(1);
     
114.    
     
115.     % Store Best Cost History
     
116.     BestCosts(it) = BestSol.Cost;
     
117.    
     
118.     % Display Iteration Information
     
119.     disp(['Iteration ' num2str(it) ': Best Cost = ' num2str(BestCosts(it))]);
     
120.    
     
121. end
     
122. 
     
123. %% Results
     
124. 
     
125. figure;
     
126. % plot(BestCosts,'LineWidth',2);
     
127. semilogy(BestCosts,'LineWidth',2);
     
128. xlabel('Iteration');
     
129. ylabel('Best Cost');
     
130. grid on;

复制代码

函数：

1. function z = Sphere(x)
   
2.     z = sum(x.^2);
   
3. end

复制代码

Molly904

回复一下,学习

Molly904

回复一下,学习1111111111

therookie

回复一下,学习

why2018470

谢谢楼主分享