An Adaptive Differential Evolution with Time-varying Inertia Weight

To enhance the search performance, the inertia weight mechanism derived from the particle swarm optimization (PSO) was employed for the differential evolution (DE). In the proposed approach, a weighted velocity was added to the trial vector to provide necessary movement, where the velocity was defined as the movement of the trial vector at the past iteration step. Consequently, consistency and diversity were introduced to the population to avoiding premature convergence. Moreover, to efficiently control the local search and convergence to the global optimum solution, time-varying inertia weight was employed. The proposed approach was employed for seven well-known benchmarks, and by comparisons with the other DE methods, the proposed approach has shown its feasibility and effectiveness.