Improved Particle Swarm Optimization Algorithm for Mixed Integer Nonlinear Programming Problems

Abstract

This paper presents an improve particle swarm optimization algorithm for solving the mixed nonlinear integer programming problems. In this algorithm, the mixed nonlinear integer programming problems is converted into unconstrained bi-objective optimization problem by the dynamic bi-objective constraint handling methods and improved the velocity equation of PSO. Introduction of migration operator in order to overcome the premature phenomenon, retention to the better performance of infeasible particles according to the constraint violation in each iteration, it is effectively maintain the swarm diversity. Numerical experiments show that the proposed algorithm has faster convergence speed and better ability of global optimization.