This document introduces computer exercises for a course on topology optimization theory and methods. It outlines the schedule and content of exercises to be completed over the course period using MATLAB codes. Students will perform basic exercises to become familiar with codes, as well as more advanced exercises involving mathematical programming techniques and multi-objective and dynamic optimization problems. The document also provides information on codes that can be used, optimization methods, filter types, optimality criteria, and practical guidance for completing the exercises.
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0 ratings0% found this document useful (0 votes)
137 views17 pages
Intro To Exercises
This document introduces computer exercises for a course on topology optimization theory and methods. It outlines the schedule and content of exercises to be completed over the course period using MATLAB codes. Students will perform basic exercises to become familiar with codes, as well as more advanced exercises involving mathematical programming techniques and multi-objective and dynamic optimization problems. The document also provides information on codes that can be used, optimization methods, filter types, optimality criteria, and practical guidance for completing the exercises.
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 17
TOPOLOGY OPTIMIZATION
THEORY, METHODS AND APPLICATIONS
INTRODUCTION TO COMPUTER EXERCISES JAKOB SNDERGAARD JENSEN 2 DTU Mechanical Engineering, Technical University of Denmark Jakob Sndergaard Jensen COMPUTER EXERCISES (the backbone of the course) Wednesday: Basic exercises with the code top.m (alternatively top88.m) Problem 1-3: Basic exercises (getting to know the code) Problem 9: Robust topology optimization Problem 12: Three dimensions Thursday & Friday: Exercises with mathematical programming Problem 4: The method of moving asymptotes (MMA) Problem 5: Mechanism synthesis using MMA Problem 6: Interior point methods (fmincon) Problem 8: min-max optimization using MMA Problem 10: SAND optimization formulation Problem 11: Multiple outputs Weekend: Catch up on missing obligatory exercises AND/OR do voluntary exercises of interest 3 DTU Mechanical Engineering, Technical University of Denmark Jakob Sndergaard Jensen COMPUTER EXERCISES Monday: Optimization in dynamics Problem 7: Optimization with harmonic loads Problem 13: Alternative measures of dynamic compliance Problem 14: Problem of your choice from Bendse & Sigmund (2004) Poster presentation of exercise work on tuesday 4 DTU Mechanical Engineering, Technical University of Denmark Jakob Sndergaard Jensen MATLAB CODES Exercises based on the 99-line code written by Ole Sigmund: CODE CAN BE DOWNLOADED FROM WWW.TOPOPT.DTU.DK 5 DTU Mechanical Engineering, Technical University of Denmark Jakob Sndergaard Jensen EXERCISES Usage: 6 DTU Mechanical Engineering, Technical University of Denmark Jakob Sndergaard Jensen EXERCISES 7 DTU Mechanical Engineering, Technical University of Denmark Jakob Sndergaard Jensen EXERCISES 8 DTU Mechanical Engineering, Technical University of Denmark Jakob Sndergaard Jensen ALTERNATIVE 88-LINE CODE (for experienced FE/Matlab people) Usage: Filter type 9 DTU Mechanical Engineering, Technical University of Denmark Jakob Sndergaard Jensen 88- VERSUS 99-LINE CODE Pros and cons of using 88-line version: Dramatically faster for large problems (speed quick-fix for 99-line code will be provided) Access to a density filter in addition to the sensitivity filter More difficult to read More difficult to modify 10 DTU Mechanical Engineering, Technical University of Denmark Jakob Sndergaard Jensen 99-LINE CODE SPEED QUICK-FIX Main problem: assembly of sparse matrix REPLACE BY: (can be downloaded on the course homepage WWW.TOPOPT.DTU.DK/DCA MM) 11 DTU Mechanical Engineering, Technical University of Denmark Jakob Sndergaard Jensen OTHER EXTENSIONS 12 DTU Mechanical Engineering, Technical University of Denmark Jakob Sndergaard Jensen FILTERS 99-line code: sensitivity filter 88-line code: sensitivity filter (ft=1) density filter (ft=2) Sensitivity filter: spatial smoothing of sensitivities to avoid checkerboards and mesh dependency - mathematically inconsistent but works well in practice Density filter: spatial smoothing of design variables + consistent update of sensitivities 13 DTU Mechanical Engineering, Technical University of Denmark Jakob Sndergaard Jensen FILTERS Implementation of sensitivity filter in 99-line code: 14 DTU Mechanical Engineering, Technical University of Denmark Jakob Sndergaard Jensen OPTIMALITY CRITERION Heuristic design updates: Design sensitivities: Adjusted until volume constraint fulfilled 15 DTU Mechanical Engineering, Technical University of Denmark Jakob Sndergaard Jensen OPTIMALITY CRITERIA Implementation: 16 DTU Mechanical Engineering, Technical University of Denmark Jakob Sndergaard Jensen PRACTICALITIES Form groups of two by the computers Get to know people ! (hook up with somebody you do not know already and preferably someone with a different background) Get User IDs and passwords from me (non-DTU students) 17 DTU Mechanical Engineering, Technical University of Denmark Jakob Sndergaard Jensen COFFEE CARDS Coffee cards will be handed out together with User IDs and passwords Prepaid 100 DKR = 25 cups of coffee, espresso, cappucino, tea, etc. IMPORTANT: return coffee cards at the end of course exercises coffee Door closes at 10pm on weekdays Open 7am-5pm in the weekend