A Particle Swarm Optimization Based on Dynamic Parameter Modification

Abstract

A new particle swarm optimization based on dynamic parameter modification is proposed in this paper (Dynamic Parameter Modification Particle Swarm Optimizer, DPSO). In DPSO algorithm , is doing oscillating decay breaking through the constraint of topical linear decreasing, and the Euclidean distance and is calculated, which respectively stand for the Euclidean distances form the position of particle to the best position that the particle has passed and the best position that all the particles have passed under the time . Parameters and of topical PSO are modified dynamically based on the comparison of and in order to coordinate between global search and local search. Then find out the optimal value of Goldstein-Price function using topical PSO and the improved DPSO respectively, and the results demonstrate that compared to topical PSO, DPSO algorithm avoids falling into the local minimum and improves the search efficiency.

Evaluation of government‐supported provincial innovation competitiveness—Evidence from China

“…Based on the data set in Table 6, perform a multi-peak function fitting for the three-dimensional model and adopt the simulation function of PSO by Zhang, Xiong, and Chen (2010).…”

An intelligent dynamic control of continuously variable transmission system using modified particle swarm optimization

“…The PSO 21–31 has a significant impact on performance of the algorithm with respect to two acceleration coefficients c1 and c2 and inertia weight γ(t). The PSO with larger inertia weight has a faster convergence speed and works well in global search, while the PSO with smaller inertia weight can reach a more accurate optimum value but only works well in local search.…”

An Improved Self-Adaptive PSO Algorithm with Detection Function for Multimodal Function Optimization Problems

“…Decreasing inertia weight particle swarm optimization is a topical algorithm, of which inertia weight decreases linearly from 0.9 to 0.4 [20][21][22]. Some scholars propose the increasing inertia weight particle swarm optimization of which inertia weight increases linearly from 0.4 to 0.9 [25]. The further development of these algorithms is fuzzy adaptive PSO, which optimizes the value of dynamically using adaptive fuzzy inertia weight controller and can solve many problems satisfactorily.…”

“…max is 6000, initial is 0.9, and final is 0.4 [25], the changes of are shown in Figure 2. The improved in (3) modifies the momentum factor dynamically in real time, while the iteration goes on and chooses the step length corresponding with the current condition according to the detection function and then achieves the self-adaptive changes according to the numerical feedback.…”