Download as PDF, PPTX
Uploaded bypaskorn
PDF, PPTX20,641 views
Simulated Binary Crossover
This document summarizes a research paper that proposes a genetic algorithm called GeneAS that can handle optimization problems with mixed variable types. GeneAS uses binary coding for discrete variables and real coding for continuous variables. It employs different genetic operators for each variable type, such as binary crossover for discrete variables and SBX crossover for continuous variables. The paper describes these genetic operators and presents results on example problems that show GeneAS finds very good optimal solutions.
![Governance, Deployment & Methodologies for Agentic Automation [2/3]](https://cdn.slidesharecdn.com/ss_thumbnails/agenticcommunityseries-day2-251112135038-a386476d-thumbnail.jpg?width=640&height=640&fit=bounds)
In this document
Powered by AI
Overview of GeneAS, an optimization method to solve engineering design problems by handling mixed variable types.
Details of design variables for cantilever beams, including materials, sizes, and dimensions, focusing on their variable types.
Introduction to the GeneAS algorithm. Strategy to manage both binary and real-coded variables in genetic operations.
Description of binary coded GA and genetic operations like crossover and mutation, with examples.
Important properties of crossover operations, including average values and spread factor implications.
Adaptations for real-coded genetic algorithms to ensure properties similar to binary-coded operations.
Implementation of SBX to maintain properties of binary crossover in real-valued variables, focusing on offspring generation.
Approach of GeneAS using different coding strategies for variables, applied separately in crossover and mutation.
Discussion of experiments comparing GeneAS to traditional methods, highlighting effectiveness in optimization.
Optimization objectives for a cantilever welded beam design and the relevant variables involved.
Summary of findings, emphasizing GeneAS effectiveness and capabilities in solving engineering optimization issues.
Further genetic operations like Blend Crossover (BLX-α) for real-parameter GAs, detailing operator performance.