A hybrid particle swarm optimizer with sine cosine acceleration coefficients

- The sine cosine acceleration coefficients (SCAC) as a new parameter adjustment strategy for the cognitive component c1 and the social component c2, respectively.
- The opposition-based learning (OBL) is adopted to initialize population.
- The sine map is utilized to adjust the inertia weight ω.
- Dynamic weight, acceleration coefficient and best-so-far position introduced to update the new position with original update formula.

Particle swarm optimization (PSO) has been widely used to solve complex global optimization tasks due to its implementation simplicity and inexpensive computational overhead. However, PSO has premature convergence, is easily trapped in the local optimum solution and is ineffective in balancing exploration and exploitation, especially in complex multi-peak search functions. To overcome the shortcomings of PSO, a hybrid particle swarm optimizer with sine cosine acceleration coefficients (H-PSO-SCAC) is proposed to solve these problems. It is verified by the application of twelve numerical optimization problems. In H-PSO-SCAC, we make the following improvements: First, we introduce sine cosine acceleration coefficients (SCAC) to efficiently control the local search and convergence to the global optimum solution. Second, opposition-based learning (OBL) is adopted to initialize the population. Additionally, we utilize a sine map to adjust the inertia weight ω. Finally, we propose a modified position update formula. Experimental results show that, in the majority of cases, the H-PSO-SCAC approach is capable of efficiently solving numerical optimization tasks and outperforms the existing similar population-based algorithms and PSO variants proposed in recent years. Therefore, the H-PSO-SCAC algorithm is successfully employed as a novel optimization strategy.