Least cost design of axially loaded compression member using particle swarm optimization

This paper gives an application of the use of particle swarm algorithm of reinforced concrete axially loaded compression members which were designed using limit state approach. The minimization of cost of these reinforced concrete compression members is carried out to get the most optimal solution in the cross section as well as in the placement of reinforcement the reinforcements at specified intervals. Particle Swarm optimization algorithm is used to find out the optimal size of the cross section and reinforcement steel. A coding has been created has been created to carry out the design for various velocity and population sizes as variable in MATLAB. The axial loads considered vary from 500kN to 2500kN. Design constraints for the design of the compression members are considered according to the Indian Standard specifications. The coding requires the user to give various input design parameters like the grades of concrete and steel, and the design live loads, percentage of steel values. The algorithm in MATLAB figures out the final e area of concrete and steel that are required for particular sections, by optimizing the overall cost of materials involved, for its construction process. Trial designs of compression members are carried out using the program and the results obtained from the non traditional technique is compared with those obtained by manual calculations by using MS Excel sheets. It is proved that the developed program is effective and adaptable and it is determined that the optimal design has been obtained by using Genetic Algorithm. From the manual designed predicted values different graphs have been drawn for different parameters whose results and equations formed will be useful for the future designers.