Enriched Particle Swarm Optimization Created By Varying Parameters for Optimization
Pooja Verma1, Raghav Prasad Parouha2

1Pooja Verma, Department of Mathematics, Indira Gandhi National Tribal University, Amarkantak Madhya Pradesh – 484887, India.
2Raghav Pasad Parouha, Department of Mathematics, Indira Gandhi National Tribal University, Amarkantak Madhya Pradesh – 484887, India. 

Manuscript received on 17 August 2019. | Revised Manuscript received on 24 August 2019. | Manuscript published on 30 September 2019. | PP: 8259-8265 | Volume-8 Issue-3 September 2019 | Retrieval Number: C6574098319/2019©BEIESP | DOI: 10.35940/ijrte.C6574.098319

Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Particle swarm optimization (PSO) is one of the most capable algorithms that reside to the swarm intelligence (SI) systems. Recently, it becomes very popular and renowned because of the easy implementation in complex/real life optimization problems. However, PSO has some observable drawbacks such as diversity maintenance, pre convergence and/or slow convergence speed. The ultimate success of PSO depends on the velocity update of the particles. Velocity has a significant dependence on its multiplied coefficient like inertia weight and acceleration factors. To increase the ability of PSO, this paper introduced an enriched PSO (namely ePSO), to solve hard optimization problems more precisely, efficiently and reliably. In ePSO novel gradually decreased inertia weight (as an alternative of a fixed constant value) and new gradually decreased and/or increased acceleration factors (meant for cognitive and social modules) is introduced. Proposed ePSO is used to solve four well known typical unconstrained benchmark functions and four complex unconstrained real life problems. The overall observation shows that proposed new algorithm ePSO is fitter than the compared algorithms significantly and statistically. Moreover, the convergence accuracy and speed of ePSO are also improved effectively.
Keywords: Unconstrained Optimization; Particle Swarm Optimization; Inertia Weight; Acceleration Coefficients.

Scope of the Article:
Discrete Optimization