The Application of an Immune Clonal Selection Algorithm Based on the Information Entropy in Truss Structure Multi-Objective Optimization

Abstract

In order to solve the conflict multi-objective optimization of truss structures between the structure minimum weight and safety redundancy, the immune clonal selection algorithm based on information entropy was adopted in this paper. Based on the immunology theory, the non-dominated neighbor-based selection, proportional cloning and elitism strategy were introduced in the multi-objective immune clonal selection algorithm (MOICSA) to enhance the diversity, the uniformity and the convergence of the obtained solution. Mathematical models for truss multi-objective optimization design are constructed, in which the information entropy value of bar stress is taken as one of objective functions, and penalty function method was used to deal with violated constraints. Several classical problems are solved using the MOICSA algorithm, and the results compared with other optimization methods. The simulation results show that the method can achieve the effect of multiple-objective optimization successfully.