【车间调度】基于遗传算法求解具有能耗约束的车间调度问题度matlab代码

✅作者简介：热爱科研的Matlab仿真开发者，修心和技术同步精进，matlab项目合作可私信。

🍎个人主页：Matlab科研工作室

🍊个人信条：格物致知。

更多Matlab仿真内容点击👇

智能优化算法       神经网络预测       雷达通信       无线传感器        电力系统

信号处理              图像处理               路径规划       元胞自动机        无人机 

⛄ 内容介绍

车间作业调度问题的优化和研究对制造企业的生产效率和生产成本有着重大的影响力,因此受到学者们的高度关注。本文在对车间调度问题方面的研究做了以下几方面的工作: 首先分析了车间作业调度问题的国内外研究现状,对车间调度问题进行了简单的描述,概述了国内外车间调度问题的研究方法。 其次介绍了遗传算法的基本理论,分析了遗传算法的特点,描述了它的应用情况。 最后在数学建模理论研究的基础上,简单分析了数学建模的方法,对数学建模常用的数学软件进行了简单的描述。根据中信戴卡轮毂制造股份有限公司某车间的车间状况,建立相应的数学模型。利用遗传算法对数学模型进行算法设计,同时利用MATLAB软件对模型求解。为公司的车间生产提供重要的帮助。

⛄ 部分代码

% RANKING.M      (RANK-based fitness assignment)

%

% This function performs ranking of individuals.

%

% Syntax:  FitnV = ranking(ObjV, RFun, SUBPOP)

%

% This function ranks individuals represented by their associated

% cost, to be *minimized*, and returns a column vector FitnV

% containing the corresponding individual fitnesses. For multiple

% subpopulations the ranking is performed separately for each

% subpopulation.

%

% Input parameters:

%    ObjV      - Column vector containing the objective values of the

%                individuals in the current population (cost values).

%    RFun      - (optional) If RFun is a scalar in [1, 2] linear ranking is

%                assumed and the scalar indicates the selective pressure.

%                If RFun is a 2 element vector:

%                RFun(1): SP - scalar indicating the selective pressure

%                RFun(2): RM - ranking method

%                         RM = 0: linear ranking

%                         RM = 1: non-linear ranking

%                If RFun is a vector with length(Rfun) > 2 it contains

%                the fitness to be assigned to each rank. It should have

%                the same length as ObjV. Usually RFun is monotonously

%                increasing.

%                If RFun is omitted or NaN, linear ranking

%                and a selective pressure of 2 are assumed.

%    SUBPOP    - (optional) Number of subpopulations

%                if omitted or NaN, 1 subpopulation is assumed

%

% Output parameters:

%    FitnV     - Column vector containing the fitness values of the

%                individuals in the current population.

%                

% Author:     Hartmut Pohlheim (Carlos Fonseca)

% History:    01.03.94     non-linear ranking

%             10.03.94     multiple populations

function FitnV = ranking(ObjV, RFun, SUBPOP);

% Identify the vector size (Nind)

   [Nind,ans] = size(ObjV);

   if nargin < 2, RFun = []; end

   if nargin > 1, if isnan(RFun), RFun = []; end, end

   if prod(size(RFun)) == 2,

      if RFun(2) == 1, NonLin = 1;

      elseif RFun(2) == 0, NonLin = 0;

      else error('Parameter for ranking method must be 0 or 1'); end

      RFun = RFun(1);

      if isnan(RFun), RFun = 2; end

   elseif prod(size(RFun)) > 2,

      if prod(size(RFun)) ~= Nind, error('ObjV and RFun disagree'); end

   end

   if nargin < 3, SUBPOP = 1; end

   if nargin > 2,

      if isempty(SUBPOP), SUBPOP = 1;

      elseif isnan(SUBPOP), SUBPOP = 1;

      elseif length(SUBPOP) ~= 1, error('SUBPOP must be a scalar'); end

   end

   if (Nind/SUBPOP) ~= fix(Nind/SUBPOP), error('ObjV and SUBPOP disagree'); end

   Nind = Nind/SUBPOP;  % Compute number of individuals per subpopulation

   

% Check ranking function and use default values if necessary

   if isempty(RFun),

      % linear ranking with selective pressure 2

         RFun = 2*[0:Nind-1]'/(Nind-1);

   elseif prod(size(RFun)) == 1

      if NonLin == 1,

         % non-linear ranking

         if RFun(1) < 1, error('Selective pressure must be greater than 1');

         elseif RFun(1) > Nind-2, error('Selective pressure too big'); end

         Root1 = roots([RFun(1)-Nind [RFun(1)*ones(1,Nind-1)]]);

         RFun = (abs(Root1(1)) * ones(Nind,1)) .^ [(0:Nind-1)'];

         RFun = RFun / sum(RFun) * Nind;

      else

         % linear ranking with SP between 1 and 2

         if (RFun(1) < 1 | RFun(1) > 2),

            error('Selective pressure for linear ranking must be between 1 and 2');

         end

         RFun = 2-RFun + 2*(RFun-1)*[0:Nind-1]'/(Nind-1);

      end

   end;

   FitnV = [];

% loop over all subpopulations

for irun = 1:SUBPOP,

   % Copy objective values of actual subpopulation

      ObjVSub = ObjV((irun-1)*Nind+1:irun*Nind);

   % Sort does not handle NaN values as required. So, find those...

      NaNix = isnan(ObjVSub);

      Validix = find(~NaNix);

   % ... and sort only numeric values (smaller is better).

      [ans,ix] = sort(-ObjVSub(Validix));

   % Now build indexing vector assuming NaN are worse than numbers,

   % (including Inf!)...

      ix = [find(NaNix) ; Validix(ix)];

   % ... and obtain a sorted version of ObjV

      Sorted = ObjVSub(ix);

   % Assign fitness according to RFun.

      i = 1;

      FitnVSub = zeros(Nind,1);

      for j = [find(Sorted(1:Nind-1) ~= Sorted(2:Nind)); Nind]',

         FitnVSub(i:j) = sum(RFun(i:j)) * ones(j-i+1,1) / (j-i+1);

         i =j+1;

      end

   % Finally, return unsorted vector.

      [ans,uix] = sort(ix);

      FitnVSub = FitnVSub(uix);

   % Add FitnVSub to FitnV

      FitnV = [FitnV; FitnVSub];

end

% End of function

⛄ 运行结果

⛄ 参考文献

[1] 栾飞,傅卫平. 改进遗传算法求解静态车间调度问题[J]. 陕西科技大学学报（自然科学版）, 2013.

[2] 常建娥, 何燕. 一种基于遗传算法求解车间调度问题的优化方法[J]. 物流科技, 2006, 29(2):2.

[3] 谢胜利, 董金祥, 黄强. 基于遗传算法的车间作业调度问题求解[J]. 计算机工程与应用, 2002, 38(10):4.

[4] 张青. 基于遗传算法的车间调度问题研究[D]. 长春理工大学.

⛳️ 代码获取关注我

❤️部分理论引用网络文献，若有侵权联系博主删除

❤️ 关注我领取海量matlab电子书和数学建模资料