Constrained TLBO algorithm for lightweight cable-stiffened scissor-like deployable structures

Present works discusses the efficient structural analysis and weight optimization of the cable-stiffened deployable structures. The stiffening effect of cables is incorporated through a matrix analysis based iterative strategy to identify the active and passive cables. The structural form can be eas...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2024-09
Main Authors: Manna, Soumyajit, Sau, Arijit, Punera, Devesh
Format: Article
Language:English
Subjects:
Algorithms
Cables
Machine learning
Matrix methods
Optimization
Penalty function
Polar coordinates
Stiffening
Structural analysis
Structural forms
Three dimensional analysis
Weight reduction
Yield criteria
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Present works discusses the efficient structural analysis and weight optimization of the cable-stiffened deployable structures. The stiffening effect of cables is incorporated through a matrix analysis based iterative strategy to identify the active and passive cables. The structural form can be easily deployed to cartesian as well as polar coordinates through the arrangement of duplet members. The large span utility of cable stiffened bar members can pose challenges to the deployability due to increased weight. A novel teaching-learning based optimization (TLBO) algorithm is utilized to optimize the overall weight of the structure through efficient section designs with proper constraint on the yield criteria. The penalty function approach is adopted to identify the unfeasible designs. A number of example cases are analysed and comparison is presented with the existing literature to show the suitability of the proposed approach. Finally, a new form of three-dimensional deployable structure is proposed. It is seen that such deployable structure can be accurately analysed using the iterative matrix analysis approach and efficiently optimized using the present algorithm.
ISSN:2331-8422