Tuning of Particle Swarm Optimization Parameter using Fuzzy Logic Sanjeev Kumar Faculty of Engineering Dayalbagh Educational Institute (Deemed Univ.) Agra, India sanjeev.85ee@gmail.com D. K. Chaturvedi Faculty of Engineering Dayalbagh Educational Institute (Deemed Univ.) Agra, India dkc.foe@gmail.com Abstract- In this paper, a fuzzy particle swarm optimization (FPSO) is developed, in which inertia weight is adaptively adjusted using fuzzy logic controller (FLC) during the search process. The FLC presented in this paper has one input and one output FLC into Particle Swarm Optimization (PSO). The effectiveness of proposed FPSO is demonstrated by applying it to four benchmark functions. The simulation result show that FPSO performs better than simple PSO. Keywords- Fuzzy Logic Controller, Inertia Weight, Particle Swarm optimization, swarm size, particle velocity. I. INTRODUCTION A new swarm intelligence technique, the Particle Swarm Optimization (PSO) is an evolutionary computation technique whose mechanics are inspired by the swarming or collaborative behavior of biological populations. The origin of the particle swarm intelligence system date back to the year 1995. It was invented by Russell C. Eberhart and James Kennedy as an optimization technique known as Particle Swarm Optimization inspired by the flocking of birds. The Particle Swarm Optimization belongs to the class of direct search methods used to find an optimal solution to an objective function/ fitness function in search space. Direct search methods are usually derivative free meaning that they depend only on the evaluation of the objective/ fitness function [1]. PSO is similar to a genetic algorithm (GA) in that the system is initialized with a population of random solutions. It is unlike a GA, each potential solution is assigned a randomized velocity and the potential solutions, called particles, are then flown through problem space. The performance of PSO can be changed by varying the values of the PSO parameters, such as: 1. Inertia Weight ω 2. Self Confidence Factor c 1 3. Swarm Confidence Factor c 2 In this paper, one input and one output FLC is introduced into the PSO. Here one input variable is iteration and output variable is inertia weight. The basic idea of utilizing this scheme is to get a better balance between exploration and exploitation during the search process. To evaluate the performance of the fuzzy PSO, a model is built to compare PSO. Section 2 gives the short description of basic PSO. In Section 3, fuzzy Particle Swarm Optimization obtained from the literature is summarizes and new fuzzy Particle Swarm Optimization (FPSO) proposed in this paper is discussed. Section 4 describes benchmark testing and the results. Finally in section 5, results are concluded. II. PARTICLE SWARM OPTIMIZATION PSO is invented by Eberhart and Kennedy in 1995. It is inspired by natural concepts such as bird flocking and fish schooling. The basic PSO algorithm consists of three steps: • Generating particle’ positions and velocities • Velocity update • Position update. First, the position, ݔ ௞ ௜ and velocities, ݒ ௞ ௜ of the initial swarm of particles are randomly generated using upper and lower bounds on the design variables values, ݔ ௠௜௡ and ݔ ௠௔௫ . The positions and velocities are given in the vector format with the subscript and superscript denoting i th particle at the time k [2]. The second step is to update the velocities of all the particles at time k+1 using particle objective or fitness values which are functions of the particles current positions in the design space at time k. Description of velocity updates and positions are 2011 International Conference on Communication Systems and Network Technologies 978-0-7695-4437-3/11 $26.00 © 2011 IEEE DOI 10.1109/CSNT.2011.44 174