Optimal Design of Roof-Truss using GA in Matlab

Volume 8 Issue 1 March - May 2019

Research Paper

Optimal Design of Roof-Truss using GA in Matlab

Krishna Amaraneni Venkata*, Chandramouli Sangamreddi**, Markandeya Raju Ponnada***

*-*** Department of Civil Engineering, MVGR College of Engineering (A), Vizianagram, India.

Venkata, K. A., Sangamreddi, C., & Ponnada, M. (2019). Optimal Design of Roof-Truss using GA in Matlab, i-manager's Journal on Structural Engineering, 8(1), 39-51. https://doi.org/10.26634/jste.8.1.15132

Abstract

An attempt has been made to design a roof truss optimally using Genetic Algorithms (GAs) and stiffness method in Matlab. The GAs tool kit functions in Matlab and the program developed in Matlab by NewCivil.com for 2D truss analysis using stiffness method are combined to design the tubular truss. The adaptive penalty function is used as the external penalty function for the constraints violation. The objective is to minimize the total weight of the truss, subjected to satisfaction of stress and displacement constraints. The cross sectional area of the truss elements is considered as decision variables. The method is tested on a 10-bar truss. The material properties of the truss such as density, Young’s modulus, allowable axial stresses (both in compression and tension) and allowable nodal displacement are taken as 2770 kg/m3, 6.89 x105 kg/cm2,1500 kg/cm2, and 5.08 cm respectively. The optimal solution with a total weight of 2472.67 kg is considered as the best in case-1 and 2288.88kg in the case-2 and the corresponding results are presented and compared with the previous research. After validation, the method is applied on a real truss. The optimal weight of the truss obtained is 138.89 kg.