New Approach in Particle Swarm Optimization by Applying Chaotic Cellular Automata-CCAPSO

Random starting point with random sequence is the nature of the Particle Swarm Optimization (PSO) as a basic parameters and capable of updating the velocity and positions of the particles by relying on this parameter. In using the PSO the main problem is in complex multi-peak search problems that usually leads to premature convergence. A new and improved method for PSO is presented in this paper by using Chaotic Cellular Automata (CCA). By using Pseudo Random Number Generator (PRNG) in PSO the proposed method is able to produce chaotic numbers for inertial coefficient (ω), acceleration sufficient (C1 and C2) and random values (rand). In the space s problem these factors leads to the appropriate random behavior for the particles and make the PSO algorithm capable of higher exploitation. Furthermore, with the combination of small steps by CCA along with unpredictable behavior in changing the inertia coefficient (ω), the proposed algorithm is able to covering the bigger problem space, avoid premature converging and falling in local minimums. This methodology shows better and faster performance in terms of searching in the space s problems in comparison with traditional algorithms.

Keywords:

PSO algorithm , Chaotic Cellular automata , Pseudo Random Number Generation , CCAPSO