Modified Asymmetric Time-varying Coefficient of Particle Swarm Optimization

In this paper, a new modification for particle swarm optimization (PSO) is developed. It has been found previously that making the learning coefficients of PSO variable enhances the performance in terms of convergence rate and obtaining the global minima solution. This has inspired a lot of researchers that investigated the effect of the coefficient's behavior on the PSO performance. However, the works in this field is still limited. This work presents a novel idea of using asymmetric curve of the modified PSO to represent the coefficient behavior. The method is tested and compared to previously reported techniques. The results are promising compared to most common methods in the field.