17-01-2013, 01:26 PM
Genetic Algorithm: A tutorial and MATLAB Program
Genetic Algorithm.docx (Size: 30.78 KB / Downloads: 27)
Abstract
— Genetic algorithms are designed to search for, emphasize optimized solutions by applying selection and crossover techniques, which are inspired by nature, to supply solutions to engineering systems. G.A are good at taking at larger, potentially huge, search spaces and navigating them looking for optimal combination of things and solutions which we might not find in a life time. This paper presents a kind of genetic algorithms in MATLAB, which have a population, selection, crossover and mutation; a researcher/student/programmer can understand easily the steps in the GA. The algorithm is simulated in MATLAB programming. A MATLAB M-code program is also provided, implementing the procedure. Objective of this note is to make researcher/student/programmer and others to interact more about the happening of the steps of the G.A, rather then any forward research work.
I. INTRODUCTION
Decision-making feature occur in all fields of human activities such as scientific and technological, and affect every sphere of our life. Engineering design, which entails sizing, dimensioning, and detailed element planning is also not exempt from its influence. Genetic Algorithms [1-2] are optimization techniques based on simulating the phenomena that takes place in the evolution of species and adapting it to an optimization problem. The idea of evolutionary computing was introduced in 1960 by I. Rechenberg in his work Evolutionary strategies. Genetic algorithms are computerized search and, optimization algorithms based on the mechanics of natural genetics and natural selection. They were proposed by Mr. John Holland based on Darwinian survival-of-the fittest principle [3], his book “Adaptation in Natural and Artificial Systems” presented the GA, which was an abstraction of natural evolution and, gave a theoretical framework for adaptation under GA. Then he presented the population-based algorithm with crossover, and the major innovation was in mutation. These techniques imply applying the laws of natural selection onto the population to achieve individuals that are better adjusted to their environment.
FUNCTIONS OF GENETIC ALGORITHM
Basic principles of genetic algorithm are: -
1. Encode the problem in binary string.
2. Random generation of population.
3. Calculate the fitness value.
4. Selection of the subjects (no auxiliary information).
5. Gene cross over and mutation.