Velocity-Driven Particle Swarm Optimization
Wei Li, Yaochi Fan, Qiaoyong Jiang, Qingzheng Xu · 2019
Particle swarm optimization (PSO) is an efficient nature-inspired optimization algorithm, which has been widely applied in many engineering fields. The performance of particle swarm optimization (PSO) has been significantly influenced by velocity update strategy. Traditionally, each particle updates its velocity based on its historical best experience and the global best experience, which may make the swarm lose its diversity and lead to premature convergence. To strengthen the performance of PSO, this paper proposes an improved PSO with the velocity-driven strategy (VD-PSO). In VD-PSO algorithm, the particles whose velocities are driven by better velocities focus on exploitation. The historical velocities, such as better velocities, may effectively characterize the landscape information of the optimization problems. In addition, the particles focus on exploration through information exchange amongst velocities. To verify the effectiveness of the proposed algorithm, experiments are conducted with CEC2014 test problems. The experimental results demonstrate the effectiveness of the proposed algorithm for solving the global optimization problems.