A nature-inspired transition from Differential Evolution to Particle Swarm Optimization

In recent years, Differential Evolution (DE) has been successfully utilized for solving multidimensional and complex optimization problems, due to its easy implementation and simplicity. Yet, similar to many other optimization algorithms, DE suffers from the problem of stagnation and lack of convergence. In contrast, Particle Swarm Optimization (PSO) has good properties in terms of its population convergence. Our proposed method is a transition from DE to PSO by using a nature inspired temperature model; same temperature model has been used in many methods such as Simulated Annealing (SA). By using our method, we benefit from both the fast speed of DE and the convergence power of PSO.