An Intelligence Artificial Fish Swarm Optimization Technique

With the massive development of information and communications technologies, the need to optimize information processing power and increase accuracy is becoming very important. This paper presents the analysis of an intelligent Artificial Fish Swarm Algorithm (AFSA) that properly select optimization parameters more effectively. It is computational intelligent with ability to solve nonlinear high dimensional problems. It addresses problems of conventional AFSA migration into local minima using control parameters such as visual distance and step sizes. Performance of the algorithm was tested using a subset of applied mathematical optimization test functions such as Ackley, Cosine Mixture, Neumaier, Rosenbrock and Rastrigin functions. Numerical results show that the intelligent algorithm outperformed the standard algorithm in 4 out of the 5 test functions. This can be very useful in computationally intensive processes.