A Comparative Study of Optimization Algorithms for Engineering Synthesis.

Abstract

A variety of optimization algorithms for engineering synthesis are included in a new general-purpose optimization computer program called ADS-1 (Automated Design Synthesis, Version 1). Preliminary testing of all presently available algorithms is conducted utilizing several carefully selected problems of significant size and complexity. These include a problem with 56 design variables and over 3500 inequality constraints. The capabilities and utility of the ADS program coupled with a structural analysis code utilizing finite element techniques is demonstrated and numerical results presented that compare the relative efficiency and reliability of the various optimization algorithms. The number of function and gradient calculations are considered important measures of merit in compring the various algorithms. A comparison of results with another existing optimization computer code is included to document the accuracy and reliability of the ADS program. Preliminary testing of the ADS program demonstrates the flexibility a design engineer would have in selecting an optimization algorithm best suited to solve a particular problem. (Author)

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 1983
Accession Number
ADA128609

Entities

People

  • Chester Michael Sprague

Organizations

  • Naval Postgraduate School

Tags

Communities of Interest

  • Ground and Sea Platforms

DTIC Thesaurus Topics

  • Algorithms
  • Coast Guard
  • Computer Programming
  • Computer Programs
  • Computers
  • Engineers
  • Evolutionary Algorithms
  • Finite Element Analysis
  • Linear Programming
  • Mathematical Programming
  • Mechanical Engineering
  • Optimization
  • Reliability
  • Structural Analysis
  • Structural Engineering
  • Three Dimensional
  • United States

Fields of Study

  • Engineering

Readers

  • Finite Element Method (FEM) for solving Partial Differential Equations (PDEs)
  • Systems Analysis and Design