Particle Swarm Optimizer with Time-Varying Parameters based on a Novel Operator
Ran Cheng, Min Yao · 2011
This paper proposes a time-varying particle swarm optimizer based on our earlier work which introduces a novel operator (leap operator). Two new parameters are recommended in leap operator to prevent premature convergence. With these two parameters, a new modification named LPSO is constructed. Since the values of the 2 parameters are not easy to determine, in this paper, they are modified as time-varying ones. With the time-varying parameters, the modified particle swarm optimizer (TVLPSO) has good potential in finding better solutions. Compared with standard PSO and LPSO, benchmark tests are implemented.