📄 nsga_2.html
字号:
<html xmlns:mwsh="http://www.mathworks.com/namespace/mcode/v1/syntaxhighlight.dtd"> <head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8"> <!--This HTML is auto-generated from an M-file.To make changes, update the M-file and republish this document. --> <title>function nsga_2(pop,gen)</title> <meta name="generator" content="MATLAB 7.0"> <meta name="date" content="2006-03-16"> <meta name="m-file" content="nsga_2"><style>body { background-color: white; margin:10px;}h1 { color: #990000; font-size: x-large;}h2 { color: #990000; font-size: medium;}p.footer { text-align: right; font-size: xx-small; font-weight: lighter; font-style: italic; color: gray;}pre.codeinput { margin-left: 30px;}span.keyword {color: #0000FF}span.comment {color: #228B22}span.string {color: #A020F0}span.untermstring {color: #B20000}span.syscmd {color: #B28C00}pre.showbuttons { margin-left: 30px; border: solid black 2px; padding: 4px; background: #EBEFF3;}pre.codeoutput { color: gray; font-style: italic;}pre.error { color: red;}/* Make the text shrink to fit narrow windows, but not stretch too far in wide windows. On Gecko-based browsers, the shrink-to-fit doesn't work. */ p,h1,h2,div { /* for MATLAB's browser */ width: 600px; /* for Mozilla, but the "width" tag overrides it anyway */ max-width: 600px; /* for IE */ width:expression(document.body.clientWidth > 620 ? "600px": "auto" );} </style></head> <body> <h1>function nsga_2(pop,gen)</h1> <introduction> <p>is a multi-objective optimization function where the input arguments are pop - Population size gen - Total number of generations</p> <p>This functions is based on evolutionary algorithm for finding the optimal solution for multiple objective i.e. pareto front for the objectives. Initially enter only the population size and the stoping criteria or the total number of generations after which the algorithm will automatically stopped. </p> <p>You will be asked to enter the number of objective functions, the number of decision variables and the range space for the decision variables. Also you will have to define your own objective funciton by editing the evaluate_objective() function. A sample objective function is described in evaluate_objective.m. Kindly make sure that the objective function which you define match the number of objectives that you have entered as well as the number of decision variables that you have entered. The decision variable space is continuous for this function, but the objective space may or may not be continuous. </p> <p>Original algorithm NSGA-II was developed by researchers in Kanpur Genetic Algorithm Labarotary and kindly visit their website for more information <a href="http://www.iitk.ac.in/kangal/">http://www.iitk.ac.in/kangal/</a></p> </introduction> <h2>Contents</h2> <div> <ul> <li><a href="#1">Simple error checking</a></li> <li><a href="#2">Objective Function</a></li> <li><a href="#3">Initialize the population</a></li> <li><a href="#4">Sort the initialized population</a></li> <li><a href="#5">Start the evolution process</a></li> <li><a href="#6">Result</a></li> <li><a href="#7">Visualize</a></li> </ul> </div> <h2>Simple error checking<a name="1"></a></h2> <p>Number of Arguments Check for the number of arguments. The two input arguments are necessary to run this function.</p><pre class="codeinput"><span class="keyword">if</span> nargin < 2 error(<span class="string">'NSGA-II: Please enter the population size and number of generations as input arguments.'</span>);<span class="keyword">end</span><span class="comment">% Both the input arguments need to of integer data type</span><span class="keyword">if</span> isnumeric(pop) == 0 || isnumeric(gen) == 0 error(<span class="string">'Both input arguments pop and gen should be integer datatype'</span>);<span class="keyword">end</span><span class="comment">% Minimum population size has to be 20 individuals</span><span class="keyword">if</span> pop < 20 error(<span class="string">'Minimum population for running this function is 20'</span>);<span class="keyword">end</span><span class="keyword">if</span> gen < 5 error(<span class="string">'Minimum number of generations is 5'</span>);<span class="keyword">end</span><span class="comment">% Make sure pop and gen are integers</span>pop = round(pop);gen = round(gen);</pre><h2>Objective Function<a name="2"></a></h2> <p>The objective function description contains information about the objective function. M is the dimension of the objective space, V is the dimension of decision variable space, min_range and max_range are the range for the variables in the decision variable space. User has to define the objective functions using the decision variables. Make sure to edit the function 'evaluate_objective' to suit your needs. </p><pre class="codeinput">[M, V, min_range, max_range] = objective_description_function();</pre><h2>Initialize the population<a name="3"></a></h2> <p>Population is initialized with random values which are within the specified range. Each chromosome consists of the decision variables. Also the value of the objective functions, rank and crowding distance information is also added to the chromosome vector but only the elements of the vector which has the decision variables are operated upon to perform the genetic operations like corssover and mutation. </p><pre class="codeinput">chromosome = initialize_variables(pop, M, V, min_range, max_range);</pre><h2>Sort the initialized population<a name="4"></a></h2> <p>Sort the population using non-domination-sort. This returns two columns for each individual which are the rank and the crowding distance corresponding to their position in the front they belong. At this stage the rank and the crowding distance for each chromosome is added to the chromosome vector for easy of computation. </p><pre class="codeinput">chromosome = non_domination_sort_mod(chromosome, M, V);</pre><h2>Start the evolution process<a name="5"></a></h2> <p>The following are performed in each generation * Select the parents which are fit for reproduction * Perfrom crossover and Mutation operator on the selected parents * Perform Selection from the parents and the offsprings * Replace the unfit individuals with the fit individuals to maintain a constant population size. </p><pre class="codeinput"><span class="keyword">for</span> i = 1 : gen <span class="comment">% Select the parents</span> <span class="comment">% Parents are selected for reproduction to generate offspring. The</span> <span class="comment">% original NSGA-II uses a binary tournament selection based on the</span> <span class="comment">% crowded-comparision operator. The arguments are</span> <span class="comment">% pool - size of the mating pool. It is common to have this to be half the</span> <span class="comment">% population size.</span> <span class="comment">% tour - Tournament size. Original NSGA-II uses a binary tournament</span> <span class="comment">% selection, but to see the effect of tournament size this is kept</span> <span class="comment">% arbitary, to be choosen by the user.</span> pool = round(pop/2); tour = 2; <span class="comment">% Selection process</span> <span class="comment">% A binary tournament selection is employed in NSGA-II. In a binary</span> <span class="comment">% tournament selection process two individuals are selected at random</span> <span class="comment">% and their fitness is compared. The individual with better fitness is</span> <span class="comment">% selcted as a parent. Tournament selection is carried out until the</span> <span class="comment">% pool size is filled. Basically a pool size is the number of parents</span> <span class="comment">% to be selected. The input arguments to the function</span> <span class="comment">% tournament_selection are chromosome, pool, tour. The function uses</span> <span class="comment">% only the information from last two elements in the chromosome vector.</span> <span class="comment">% The last element has the crowding distance information while the</span> <span class="comment">% penultimate element has the rank information. Selection is based on</span> <span class="comment">% rank and if individuals with same rank are encountered, crowding</span> <span class="comment">% distance is compared. A lower rank and higher crowding distance is</span> <span class="comment">% the selection criteria.</span> parent_chromosome = tournament_selection(chromosome, pool, tour); <span class="comment">% Perfrom crossover and Mutation operator</span> <span class="comment">% The original NSGA-II algorithm uses Simulated Binary Crossover (SBX) and</span> <span class="comment">% Polynomial mutation. Crossover probability pc = 0.9 and mutation</span> <span class="comment">% probability is pm = 1/n, where n is the number of decision variables.</span> <span class="comment">% Both real-coded GA and binary-coded GA are implemented in the original</span> <span class="comment">% algorithm, while in this program only the real-coded GA is considered.</span> <span class="comment">% The distribution indeices for crossover and mutation operators as mu = 20</span> <span class="comment">% and mum = 20 respectively.</span> mu = 20; mum = 20; offspring_chromosome = <span class="keyword">...</span> genetic_operator(parent_chromosome, <span class="keyword">...</span> M, V, mu, mum, min_range, max_range); <span class="comment">% Intermediate population</span> <span class="comment">% Intermediate population is the combined population of parents and</span> <span class="comment">% offsprings of the current generation. The population size is two</span> <span class="comment">% times the initial population.</span> [main_pop,temp] = size(chromosome); [offspring_pop,temp] = size(offspring_chromosome); <span class="comment">% temp is a dummy variable.</span> clear <span class="string">temp</span> <span class="comment">% intermediate_chromosome is a concatenation of current population and</span> <span class="comment">% the offspring population.</span> intermediate_chromosome(1:main_pop,:) = chromosome; intermediate_chromosome(main_pop + 1 : main_pop + offspring_pop,1 : M+V) = <span class="keyword">...</span> offspring_chromosome; <span class="comment">% Non-domination-sort of intermediate population</span> <span class="comment">% The intermediate population is sorted again based on non-domination sort</span> <span class="comment">% before the replacement operator is performed on the intermediate</span> <span class="comment">% population.</span> intermediate_chromosome = <span class="keyword">...</span> non_domination_sort_mod(intermediate_chromosome, M, V); <span class="comment">% Perform Selection</span>⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -