Scribd
Upload
33 views

Real Coded Genetic Algorithm

This document outlines the process for a real coded genetic algorithm. It specifies the following parameters that must be initialized: population size, crossover probability, mutation probability, distribution indexes for crossover and mutation, number of variables, variable bounds, and number of iterations. The pseudocode then describes the genetic algorithm process of initializing a random population, evaluating fitness, performing tournament selection and crossover or mutation to generate offspring, evaluating offspring fitness, and combining the population and offspring. The document also provides details on the SBX crossover and polynomial mutation operators used to generate new solutions in the real-coded genetic algorithm.

Uploaded by

SURABHI SHARMA
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
33 views

Real Coded Genetic Algorithm

This document outlines the process for a real coded genetic algorithm. It specifies the following parameters that must be initialized: population size, crossover probability, mutation probability, distribution indexes for crossover and mutation, number of variables, variable bounds, and number of iterations. The pseudocode then describes the genetic algorithm process of initializing a random population, evaluating fitness, performing tournament selection and crossover or mutation to generate offspring, evaluating offspring fitness, and combining the population and offspring. The document also provides details on the SBX crossover and polynomial mutation operators used to generate new solutions in the real-coded genetic algorithm.

Uploaded by

SURABHI SHARMA
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 8

REAL CODED GENETIC

ALGORITHM

Submitted to : Mr. Sudip K. Ganguly


Submitted by : Surabhi Sharma
ID: 2017A1PS0682H
Parameters to be specified initially:

 Population Size, Np
 Crossover probability, pc
 Mutation probability, pm
 Distribution index for crossover, etac
 Distribution index for mutation, etam
 Number of variables, D
 Lower and upper bounds for each variable, lb(1:D) & ub(1:D)
 Number of iterations, T
PSEUDOCODE for real coded GA

INPUT: Fitness function, lb, ub, Np, T, Pc, Pm, ŋc, ŋm, k
1. Initialize a random population, P
2. Evaluate fitness(f) of P
for t = 1 to T
Perform Tournament selection of tournament size, k
for i = 1 to Np/2
Randomly choose two parents
if r<Pc
Generate two offspring using SBX-Crossover
Bound the offspring
else
Copy the selected parent as offspring
end
end
Pseudocode continued..

for i = 1 to Np
if r<Pm
Perform polynomial mutation of ith offspring
Bound the mutated offspring
else
No change in the ith offspring
end
end
Evaluate the fitness of offspring
Combine population and offspring to perform (mu + lambda) strategy.
end
SBX Crossover(if rc<pc)
 Generate a random number ‘u(j)’ for each variable ‘j’ that needs to be optimized.
 Compute beta (for each variable):

 According to the different beta values obtained for each variable ‘j’, if Pa’ = Parent 1 & Pb’ = Parent 2, then Offspring 1 and Offspring 2
are to be calculated as:

where
Oa = Offspring1
Ob = Offspring 2
The two offspring are symmetric about the parents to avoid any bias.
Polynomial Mutation(if rm<pm)

 Generate a random number ‘r(j)’ for each variable ‘j’ that needs to be optimized.
 Compute delta (for each variable):

 For each variable ‘j’, the new offspring O is to be calculated as:

Taking different delta, upper bound and lower bound values for each variable one by one.
 When r = 0, delta = -1 & when r = 1, delta = 1; these are respectively the boundary values for delta. Hence delta
ranges from [-1,1].
THANK YOU!
Important aspects to remember

You might also like